Scientists Worry Machines May Outsmart Man

Strudelkugel writes "The NY Times has an article about a conference during which the potential dangers of machine intelligence were discussed. 'Impressed and alarmed by advances in artificial intelligence, a group of computer scientists is debating whether there should be limits on research that might lead to loss of human control over computer-based systems that carry a growing share of society's workload, from waging war to chatting with customers on the phone. Their concern is that further advances could create profound social disruptions and even have dangerous consequences.' The money quote: 'Something new has taken place in the past five to eight years,' Dr. Horvitz said. 'Technologists are replacing religion, and their ideas are resonating in some ways with the same idea of the Rapture.'"

Comments

I thought this was the whole point?

[Anonymous Coward]

Religious people are regularly and strongly ridiculed.

50% of scientists call themselves Democrats, 48% call themselves Independents, 2% calls themselves Republicans - in a recent poll described on Slashdot.

I kind of thought the development they describe was the goal all along.

Re:I thought this was the whole point?

[Devout_IPUite]

Regardless of political orientation, this research WILL get done. If the US doesn't get it done, China will. How does that make you feel?

Re:I thought this was the whole point?

[Anonymous Coward]

Hungry? And maybe a little excited? But maybe that's just my normal morning feelings getting mixed up...

Re:I thought this was the whole point?

[Devout_IPUite]

Also from the article "The researchers also discussed possible threats to human jobs, like self-driving cars, software-based personal assistants and service robots in the home. Just last month, a service robot developed by Willow Garage in Silicon Valley proved it could navigate the real world."

An interesting thing to note is this: When a computer exists that is as intelligent as a stupid human, almost every job at and close to minimum wage vanishes. Robots can and will get cheaper than a human worker, no one will need taxi cab drivers, grocery store baggers, first tier phone customer service reps, construction workers, janitors, garbage men, delivery men, mail men, traffic cops, book keepers, data entry people, secretaries, fast food chefs, etc.

At this point we will have two choices as a society. 1) Let them (the stupid people) starve, 2) give them welfare for no other reason than they're economically useless.

Re:I thought this was the whole point?

[hanabal]

have you thought about the posibility that when robots do all the jobs that no one wants to do, productivity might increase by enough to allow all the people to live comfortably. Also I don't think that valuing people only by their economic worth is very nice.

Re:I thought this was the whole point?

[knarf]

You forgot to mention 'programmers'... a whole section of the /. population would be out of work. The mean task of turning structural drawings into physical or logical reality is something which computers will be able to do far more efficiently than humans. Programmers are construction workers of logic instead of wood, steel and concrete. Architects might survive a bit longer before they, too, are made redundant.

Re:I thought this was the whole point?

[OeLeWaPpErKe]

have you thought about the posibility that when robots do all the jobs that no one wants to do, productivity might increase by enough to allow all the people to live comfortably. Also I don't think that valuing people only by their economic worth is very nice.

Do tell, there is this theory that has a few things to say about what happens when the economy is larger than the needs of it's population. It's generally referred to as evolution. What exactly does it say will happen in that case ?

Hint : it has something to do with population growth ...

Re:I thought this was the whole point?

[toriver]

Stop drinking the Model-Driven Development Kool-Aid.

Re:I thought this was the whole point?

[billcopc]

Ideally there should be another choice: 3) send the dumb ones back to school.

We all know that is not going to happen because:

1. they don't wanna go to school in the first place
2. the educational system in its current state is not economically viable for these people (nor the society actually footing the bill)
3. like any parasite, they will get together and lobby for free handouts while opposing progress, like they have always done (churches, exclusive communities, 3rd world expats)

The fact of that matter is that at some point in the not-so-distant future, there will be some hard sacrifices to be made if we want to improve the quality of life on our little blue planet. The problem is no one wants to "play god" because of the unpredictable consequences of sending a large number of arbitrarily selected people to an early grave. Humans are selfish by nature and we are not willing to sacrifice our own well-being for that of another.

Re:I thought this was the whole point?

[Devout_IPUite]

You sound like a socialist. That's fine, I'm actually a socialist too. It's still going to be a huge debate (at least in my country) if we should let them starve or give them free food, health care, housing, and education.

Certainly for the intelligent people, having a swarm of stupid lackeys would increase productivity. But the stupid people are still going to take more energy to run than a computer after one or two dozen generations (humans baseline energy consumption is around 72w, you can get computers that run on 5 watts http://www.linuxfordevices.com/c/a/News/100-Linux-wallwart-launches/).

Re:I thought this was the whole point?

[Devout_IPUite]

The programmers will be safe for a few machine generations past the grocery store baggers I suspect. It's quite possible that the accountants, studio musicians, programmers, carpenters, and such finding themselves without jobs will be the catalyst to turn us into socialists.

Re:I thought this was the whole point?

[Devout_IPUite]

Frankly, many people at and near minimum wage would not be useful if they managed to eek out an engineering degree (they'd still be terrible engineers, the kind that everyone realizes was actually harming overall productivity once they're gone).

Re:I thought this was the whole point?

[noname444]

What's wrong with having robots do boring tasks for us while we're doing something fun and/or useful? Hell, isn't that the whole point?

Re:I thought this was the whole point?

[oneirophrenos]

1) Let them (the stupid people) starve

They are not going to starve. If there's one thing to learn from poverty, it's that it makes people revolt and rebel. Welfare is a means with which to pacify the poor so you'll have at least some form of social order in a society where unemployment exists.

Re:I thought this was the whole point?

[maxwell demon]

Ideally there should be another choice: 3) send the dumb ones back to school.

We all know that is not going to happen because:

1. they don't wanna go to school in the first place

That's not really a problem: Most of them don't like to do the jobs they do either. But they do them anyway, because they need the money.

Make schools pay money to the people going there (depending on how well they do), and the people will go to school and learn. OK, there's the problem of where that money should come from. Well, simple: From those who profit from having more well-educated people around.

2. the educational system in its current state is not economically viable for these people (nor the society actually footing the bill)

Fix it. Yes, I know that's easier said than done :-)

3. like any parasite, they will get together and lobby for free handouts while opposing progress, like they have always done (churches, exclusive communities, 3rd world expats)

You are not really unbiased, are you?

Re:I thought this was the whole point?

[unlametheweak]

have you thought about the posibility that when robots do all the jobs that no one wants to do, productivity might increase by enough to allow all the people to live comfortably.

That's only in places where the equitable distribution of wealth is valued and regulated by legislation. In a place like America this would be considered "communism". Everywhere else it is just Utopian thinking because technology has rarely if ever made work easier on people, it just enabled people to work harder. So, people may get their smart phones in the Utopia of the Future, but they will also be required to be on call 24/7.

Re:I thought this was the whole point?

[pizzach]

Makes me feel a sudden need to add a provisional law to cut down on the Chinese Robot Immigration....

Re:I thought this was the whole point?

[Anonymous Coward]

we have more options:
3) make them lab rats (I am not supporting this but someone might)
4) teach them skills needed for upcoming years. well, with singularity so near it might not make sense but we can still give it a try.
5) there are always jobs that involve more than simple thoughtless work.
6) support one or two generations on welfare while switching to genetically engineered babies. people can still choose to procreate in the natural way but they should be leave enough fortune just in case the young one cannot make living on its own.

Re:I thought this was the whole point?

[Anonymous Coward]

I'm sorry, but...

NAZI! Bloody Nazi! We're not supposed to invoke goodwins law but screw that! There's a bloody Nazi post winking me in the face! I want to print it out, cut it up and make some Nazi Goreng!

Re:I thought this was the whole point?

[Trahloc]

Perhaps, but that wallwart doesn't look nearly as good in lingerie... but if cherry2000 is any indication they'll be solutions for that as well... although I think they'll need more than 5 watts to run.

Re:I thought this was the whole point?

[Trahloc]

As I am on call 24/7 I can honestly say I prefer it to having to work the fields from sunup to sundown like my grand parents.

Re:I thought this was the whole point?

[Frogg]

we need to stop looking at unemployment as being a problem...

i think it may have been Robert Anton Wilson who said "unemployment is a benefit of a technologically advanced society" - and i have to agree with that view really. afterall, we are always inventing 'labour saving devices' - and this is really just an extreme extension of that, indeed perhaps one should say it's the ultimate extension of that. i believe we will eventually replace most human work (whether it be thinking-based or labour-based work) with that of machines. (sure, it may be some time off, and some countries may do it before others of course, but it will happen eventually)

how we go about 'solving' this issue exactly, with 'welfare' or 'benefits' (as the system is called here in the UK) is still an unanswered question. (and maybe the AIs will come up with a solution for that too, given the right data and a bit of time to 'think' about it)

fwiw i think that that view of unemployment as being a problem is deep rooted - take for example yourself, your view expressed above fails to encompass the fact that eventually /you too/ will probably be out of work eventually, alongside 'them (the stupid people)'.

we /all/ need to think differently with regards to this, because if AI begins to take-off in this manner then it probably won't be very long before we're /all/ shown as being less intelligent than the machine AIs - it won't be very long before that bar is raised.

it's not like there's much fundamental difference between those who are 'stupid' and have to do the lower jobs of society and those who are 'intelligent' and get more choice about what they are able to do regarding employment/work/jobs - we're all human, and mostly just the victims of circumstance, and the education that arose because of those circumstances.

i, for one, welcome our up-and-coming AI overlords!

(what's going on with slashdot these days?!? - i really hope my comment has better formatting than any of these previews seem to show - i've even chosen html now, and added p-tags around all my paragraphs, and /still/ it looks a state in the preview view [sigh])

Re:I thought this was the whole point?

[Xeth]

I think you overestimate the number of people that would be subject to that kind of reasoning. How many programmers are given the task of simply implementing absolutely complete and logically consistent specifications?

Re:I thought this was the whole point?

[Rocketship Underpant]

Luddites have said something similar for every invention since the cotton gin... and probably since fire was discovered, for that matter. The fact is that such fantastic robots would make the economy so productive we'd hardly need to work more than a few hours a week regardless of our skills. At any rate, I doubt the AI singularity is just around the corner, whatever its repercussions may be.

Re:I thought this was the whole point?

[Xeth]

Addendum: This is particularly obvious when you consider modern high-level languages. If you're going to fully-specify something in pseudo-code, you could probably just give it consistent syntax and call it Python.

Re:I thought this was the whole point?

[VJ42]

At this point we will have two choices as a society.

This isn't the first time we've had to make this type of choice, previously we had the Industrial Revolution, when mechanisation made the jobs of many those in agriculture, transport, manufacturing etc. redundant. those who opposed it aren't remembered kindly. Indeed, the "Luddite fallacy" is well understood by economists.

Re:I thought this was the whole point?

[kz45]

"have you thought about the posibility that when robots do all the jobs that no one wants to do, productivity might increase by enough to allow all the people to live comfortably. Also I don't think that valuing people only by their economic worth is very nice."

It may not be "nice", but it needs to be discussed. If you have no education or skill, how will you make money if a robot can do the exact same thing? It's not like the need for money is going to magically disappear.

Even if this does happen, I would imagine the unions would have their say..and force companies to hire people anyway, even if it isn't necessary.

Re:I thought this was the whole point?

[Anonymous Coward]

I can't wait to get hold of the new perpetual beta of Google God 2.0. Now with 20% fewer apostles! Google God feeds the need for a creed with more speed! After all, when you do no evil you are close to divine. Let Google define your divinity, divine your trinity, and combine your affinity.

Belief is just a click away. Ads completely tailored to your new Google faith.

Because, you know, in Soviet Russia evil does you.

Re:I thought this was the whole point?

[Smallpond]

Don't worry, there will always be a need for skilled typists, file clerks, elevator attendants, telephone operators and musicians to accompany the silent movies.

Re:I thought this was the whole point?

[Philip K Dickhead]

Two words: Artificial Flowers. :-)

And that is why the paranoid fantasy of a world run by Lesbian Rapist Robots will never occur.

Re:I thought this was the whole point?

[rhoderickj]

That's a pretty ignorant viewpoint. If you think that people in service-based industries are all just "stupid," then you are exhibiting nothing but arrogance and naivete. I bet you're one of these assholes aren't you?

Re:I thought this was the whole point?

[nine-times]

I assume you're trying to be funny, but I have a couple objections here:

First, what makes you so sure that service reps, construction workers, and traffic cops are all stupid? It's true that some of these people might not have very intellectually taxing jobs, but that might not be the extent of their ability. Einstein was just a patent clerk, after all. But also, some of these jobs do take some intelligence. For example, a "construction worker" might not be using his head too much if he's sweeping up trash, but at a certain level, you need a certain understanding of physics and engineering to do good carpentry.

And what do you do that's so smart? I've known people in IT, both on the support and coding side, who were relative morons. What if AI someday handles those jobs too? Are you sure that you won't be counted among the "stupid people"?

My second problem is this idea of letting people starve or "giving them welfare". If we ever really get to the point where robots/AI can do most of the work for us, and no other new work shows up as being necessary, then won't that completely reshape the economic landscape? I'm not sure "giving people welfare" will make a lot of sense in that context, given that we should all be living lives of leisure at a minimal cost.

I anticipate someone saying, "well, no, because resources will still be limited, and there won't be enough robots to go around." Ah, so then robots still won't be able to do everything for us, and we'll need people to do the remaining work. Looks like we have jobs again.

And there's the problem with your notion of "Let them (the stupid people) starve". What makes you think the stupid people won't all revolt at that point? Or assuming they don't revolt, why wouldn't those stupid people get to work providing for themselves? I mean, if they have no food because they have no jobs, then won't they also have all day free to find ways of getting food? Again, you have work.

To the extent that your post is serious, it shows a serious lack of understanding.

Re:I thought this was the whole point?

[eltaco]

it's not just not very nice - it's social darwinism.
GP post is right on the money (apart from their last paragraph) - it's called the third industrial revolution and it's been making people unemployed since the 80s.
competition forces companies to eventually lower their costs. with robots and computers being able to do more and more human jobs, it seems like a good idea to fire workers and have them replaced.
on the surface it seems like a good idea - but high unemployment, which eventually follows, has never been good for any economy.
it won't bring on a new era of prosperity, as less people will be able to buy their products. this forces companies to lower prices even more (ie firing workers, using technology instead), which again hurts purchasing power. A lovely vicious circle ending in the very rich getting richer and society's bottom 50% starving.

you're correct that a free workforce can heighten productivity immensly. but that doesn't fly in our current economic model. when using (robotic) slaves, it has only ever truly benefit the rich.

Re:I thought this was the whole point?

[Philip K Dickhead]

OK Mr. Malthus.

Murder by numbers,
1,2,3,
It's as easy to do,
As your ABC...

First of all, your assumption that it is stupid people who do simple labour - rather than the socially marginalized - is absurd, offensive and not worthy of deeper critical examination, except by way of devastating the thought.

Your proposition is "Santa Claus" economics - If you have something, it must be because you deserved it and if you are in poverty of opportunity and money? You deserved that, too.

That's how slow genocide has been perpetrated against the native populations of United States, Australia and Southern Africa.

I have had my own shoes shined, and been driven in cabs by people who's bags I am not fit to carry - by means of either their intellect or simple good will and sheer humanity.

But it is clear that valuing humanity would be a difficult conception for you.

 

Re:I thought this was the whole point?

[mabhatter654]

This will apply in countries that strictly limit their immigration and are already declining in population. .think Japan and Scandinavia. They are riding the curve perfectly so "menial" laborers will be few enough that they'll all be "robot repairmen". As the productivity burden combine with the top heavy portion of retirees will require workers to support 3-4 other families from the value of their labor. But it only works when immigration is tightly controlled.

It won't work in the US because we allow lots of off the books immigration that reduces the need for "labor saving devices" when we can have non-citizens (non-rights holders) do the work for cheaper.

Valuing people only by their economic worth is something communists do... it was one of the reasons capitalism won while communist workers were unmotivated and non-productive.

Re:I thought this was the whole point?

[hughbar]

Well, the winners write history, that's one sure thing. But the industrial revolution wasn't exactly an unconditional blessing, then or now.

Secondly, none of those thing were choices then and they are not now. They're usually candy-wrapped economic coercion in a state that Guy Debord calls Augmented Survival: http://www.bopsecrets.org/SI/debord/2.htm.

So, it's preferable to go down this road to deliver universal plenty, but given our unreasoned, current economic orthodoxies, that's unlikely to happen.

Re:I thought this was the whole point?

[LinuxDon]

There will always be things that computers can't do.
Think about the time when all of the people that were typing invoices all day on a typewriter lost their job.
They all found a new job.

Also, who will maintain all of these new systems (which will likely be very prone to failure because of their complexity)?
It will generate a lot of maintenance jobs. A lot of this work will be routine work which can be carried out by almost any of the people you mentioned.

Re:I thought this was the whole point?

[Philip K Dickhead]

Other than the Silent Movie crack, the elimination of these type of positions has been done not to make a more convenient world for most - but to line the pockets of your trillionaire, super-rich class, at the expense of the livelihood of large segments of the society, and to the detriment of the general humanity of life in the 'modern' age.

Perhaps a machine cocoon life - the logical conclusion of your argument - is your objective? Never meet another person, as long as you live?

Sealed in the robo-cocoon, all your needs met by servile technology. Fluids and solids delivered and extracted from various orifices, in your cyber-slumber. Welcome to the Blue-Pill universe.

"What floor, please?" is an opportunity to interact as a human being, in some small way. The disappearance of this from the world is a little, incremental darkening... a diminution of the real quality of life.

But you do you care about humanity? You can eat Cheetos and fuck a robot.

You were fooled, with every one else, that you live in an age of "progress". The abstracted, technological form of human predation that is this world is no progress.

Re:I thought this was the whole point?

[mabhatter654]

historically, well-to-do states self-limit the birth rate because of economic selfishness. Look at Japan or Scandinavia... They have just 1-2 children (from 2 adults) so that's negative growth. They live a long time, and the children are highly schooled, and well cared for... unlike in India where you have to have 4-5 kids just to make sure a few live to be productive adults so they can take care of you. Also, the strong social programs (medical care, pensions, etc) reduce the need to have kids as economic "insurance", so they're actually a liability in terms of costs to feed, clothe, school, free time, social calender, etc. Rich people have fewer children because it distracts from making money and doing what they want!

Even in the US, the birth rate from non-immigrant citizens is already negative. Growth comes mostly from all the students and workers we import that still have the old views of children for economic reasons.

Re:I thought this was the whole point?

[canadian_right]

In 1831 90% of the USA population worked on farms. Today that is less than 2%. As technology improved the number of people required to produce food has greatly diminished, and people were talking about the same problem these robots and AI's might cause. What would all the farm workers do! Most went to factories then the service sector.

The same story is told about virtually all technological progress, from seamstresses rioting over sewing machines, water powered mills, the steam engine, and the modern factory displacing cottage industry, pundits have shouted that there will be widespread unemployment, riots, pandering, and society will collapse!

They were all wrong.

Big changes do cause short term upheavals, and a truly intelligent AI mated with a general purpose robot will cause huge changes to society, but these changes will free people from boring manual labour to do more creative work. And the non-creative? They'll do their one days work, or one hour, or none, then watch tv just like they do now.

The 5 day work week was a radical change. Eventually technology will bring us the one day work week then no work. Trying to ban technology won't stop it. Society will be greatly different. I think overall people will be more free and happier when we live in a post-scarcity society.

Re:I thought this was the whole point?

[scottzak]

have you thought about the posibility that when robots do all the jobs that no one wants to do, productivity might increase by enough to allow all the people to live comfortably

Not to quibble, but all people could live comfortably on what we produce now. There is, unfortunately, the political problem of many people wanting to live lavishly, and a significant subset of those who are determined to live obscenely.

Re:I thought this was the whole point?

[init100]

In a place like America this would be considered "communism".

And thus evil. Letting people starve on the other hand, would be considered noble.

Re:I thought this was the whole point?

[Memroid]

I think starvation would be the best choice; otherwise, being a couch potato becomes the most popular career path among college students everywhere.

Re:I thought this was the whole point?

[Anonymous Coward]

Wow, that last name in your UID is extremely correct.

Re:I thought this was the whole point?

[NeverVotedBush]

And that is why the paranoid fantasy of a world run by Lesbian Rapist Robots will never occur.

Damn. I was so hoping...

Re:I thought this was the whole point?

[ChronoFish]

"...An interesting thing to note is this: When a computer exists that is as intelligent as a stupid human, almost every job at and close to minimum wage vanishes..."

While it may seem "obvious" this is not correct. There has to be cost benefit.

I work in a medical lab - you'd *think* that it would be more cost effective to employ robots to handle cups of "body fluids" - and in some cases it is. But as of yet, we have a lot more people than robots, not because the robots aren't capable, but because they are just too damn expensive for the volume of cups we process.

A second follow up to your post is that "minimum wage" jobs are not the only ones targeted. In fact, again in our case, the more expensive a job is, the more likely that job is to be replaced by automation when the automation is available.

We have two labs - one which requires complex sample prep. This takes an educated person many steps. "Educated" = money and "many steps" = time and together equals "lots of money" - and has been the first area targeted for automation. The second lab does not require a 4-year degree, and the sample prep is about as difficult as data-entry and pouring from a cup to a tube. Here the economics are such that it's *better* to have hired help rather than robotics.

My final point:
Robots break. When robots break everything halts. This is immensely expensive from both loss of productivity and the repair itself. By contrast our man-operated lab can always do "something" even if the electricity goes out or the computer network goes down. Humans are much more adaptable that way (though they do tend to bitch and moan more).

-CF

Re:I thought this was the whole point?

[FatdogHaiku]

Well, first off people that DON'T have low level jobs will have to get a lot better at what they do. There will be no carpenter with 20 years experience to adapt the work to the inadequacies of the architect (happens all the time, they permit the changes later using the term "as built"). Also, I would think that because the first units are going to be expensive, they will target more expensive workers. You don't outsource the cheap labor... you outsource the high cost stuff...

I get this vision of "5" pecking away at a keyboard, working out his first "Hello World"...

Re:I thought this was the whole point?

[brentonboy]

An interesting thing to note is this: When a computer exists that is as intelligent as a stupid human, almost every job at and close to minimum wage vanishes. Robots can and will get cheaper than a human worker, no one will need taxi cab drivers, grocery store baggers, first tier phone customer service reps, construction workers, janitors, garbage men, delivery men, mail men, traffic cops, book keepers, data entry people, secretaries, fast food chefs, etc.

At this point we will have two choices as a society. 1) Let them (the stupid people) starve, 2) give them welfare for no other reason than they're economically useless.

Yah, that's a Good Thing. The cost of an autonomous garbage pick up system would presumably be much cheaper in the long run, and we will be free to use that money on things we want to, like going to the movies. So the garbage man will be able to quit the job he hates, but took because he needed the money, and he'll be able to get a job perusing his real interests of film making or perhaps robotics.

The concern for "jobs" is idiotic and short sighted. Most people work boring and trivial jobs which a robot could perform not because it's their passion, but because they need money. If these jobs can be automated, that will open up the opportunity for more jobs that people actually enjoy doing, like planning out where to set up a robitically controlled broccoli farm, choosing the location, negotiating price, programming the computer that manages it, purchasing the hardware, etc... That sounds way more fun to me than driving a combine back and fourth across a field for hours each day.

Re:I thought this was the whole point?

[OeLeWaPpErKe]

So evolution is going to magically reverse on itself just when it serves our purpose ?

If any group chooses to limit it's birthrate artificially they will soon find themselves replaced by another group who chooses not to do so - unless external factors intervene (ie. discrimination between those groups, and since it's mostly ethnic differences between such groups, racism).

This happens at an astonishing rate. Suppose population is divided 90%-10%. Suppose also that the majority has a lower birthrate (1.5 per woman) than the minority (2.5 per woman) (and suppose parents die when they've had their kids, and these kids are all born at the same time, and that that time is 25 years after birth). It takes less than 5 generations for the minority to become the majority. A little over a century. The generation after that, said minority is 2/3rds of the population, next generation it is over 80%. And the 8th generation the minority has over 95% of all the population.

Re:I thought this was the whole point?

[NeverVotedBush]

have you thought about the posibility that when robots do all the jobs that no one wants to do, productivity might increase by enough to allow all the people to live comfortably. Sure, it might happen, but listen for the deafening roar when people line up to decry such a plan as taking care of the lazy and non-productive.

But we could retrain them! Educate them!

And that is happening how much now?

Re:I thought this was the whole point?

[Ihmhi]

It's not just "stupid" people who work jobs like grocery baggers. Don't forget about teenagers (goodbye extra college money/car money!), the elderly, the disabled, etc.

Re:I thought this was the whole point?

[Dr_Barnowl]

Productivity is already high enough for everyone to live comfortably, and has been for some time. In America, since 1983, the bottom 80% of the population have had less than 20% of the wealth.

Re:I thought this was the whole point?

[Dr_Barnowl]

Ah, but for the hours you'd use it, it would be 5 watts on average. And it wouldn't demand to watch Jane Austen movies and have the house redecorated.

Re:I thought this was the whole point?

[greg_barton]

Not only that, but people would have more time to educate themselves and become economically viable.

Re:I thought this was the whole point?

[unlametheweak]

have you thought about the posibility that when robots do all the jobs that no one wants to do, productivity might increase by enough to allow all the people to live comfortably

And I suppose I should add, that's what we have migrant labour for. Of course there will always be jobs that people don't want to do, and the productivity gains will be pocketed by the owners of the means of production. I'll be laughing when the prison population of the United States starts to saturate at 50% of the population. Labour is most affordable to corporations in U.S. penitentiaries, so I suspect the trends to continue.

Re:I thought this was the whole point?

[m.ducharme]

" A few machine generations..."

That should take what, two, three days?

Re:I thought this was the whole point?

[AnyoneEB]

See Manna for one sci-fi view of what happens when robots take over all of the jobs. It is supposed to be realistic, but I think it, like a lot of near future sci-fi, overestimates the speed of technological progress.

Re:I thought this was the whole point?

[KahabutDieDrake]

Letting them starve wouldn't be noble here in America (USA?). However, opening a hunting season to "thin the heard" would be. It's already done with lesser species, why not lesser people?

Disclaimer, I don't hunt except for food. Human tastes like chicken.

Re:I thought this was the whole point?

[Beefpatrol]

I had a similar discussion with a friend a few weeks ago about something similar. We were not talking about AI taking over the unskilled jobs; we were talking about the rather tight coupling between full time corporate employment and health insurance in the US. My contention was that having those uncoupled would allow much greater economic flexibility and production efficiency for the country because the risk of leaving a corporate job and starting a venture would be greatly reduced if comparable independant health insurance was affordable. The increased production efficiency in this case would come from a better match between an individual's money making activities and the individual's strengths and preferences. Just about everyone works better and harder at things they like than things they dislike, and being good at something generally makes someone more efficient. If one could get affordable health insurance without having to work full time for an employer that offers health care, one could start one's own venture doing something that matches one's strengths and preferences. Hopefully this would cause an increase in production and efficiency.

If that effect were real, the same could be true in the case of a society where the crappy brainless jobs are done by automatons of some sort. People thought the same thing, however, back when things like washing machines and dishwashers were becoming a common part of American life. They also said the same thing when computers were becoming common in business. Even if one were to correct for the relatively poor match between people's desires and skills and the capabilities and requirements that business software often involved, I'm not sure that people became more efficient or more self-actualized as a result of having computers at their disposal. It appears that jobs that a computer could do simply went away, and the other job descriptions just expanded to include the operation of a computer to do the job that used to be done by a person. Required hours went up, required qualifications went up, productivity went up, and the people whose jobs went away either retrained for something else or retired or had financial problems. In the case of washing machines, those who used to wash clothes by hand ended up making horrific casseroles involving jello and weiners, and that *obviously* wasn't an improvement. Any way you look at it, replacing unskilled labor with machines will probably make society more productive, but without education for those who no longer have a job, society will end up with lots of hobos and laughably bad food.

Re:I thought this was the whole point?

[grolaw]

Also from the article "The researchers also discussed possible threats to human jobs, like self-driving cars, software-based personal assistants and service robots in the home. Just last month, a service robot developed by Willow Garage in Silicon Valley proved it could navigate the real world."

An interesting thing to note is this: When a computer exists that is as intelligent as a stupid human, almost every job at and close to minimum wage vanishes. Robots can and will get cheaper than a human worker, no one will need taxi cab drivers, grocery store baggers, first tier phone customer service reps, construction workers, janitors, garbage men, delivery men, mail men, traffic cops, book keepers, data entry people, secretaries, fast food chefs, etc.

At this point we will have two choices as a society. 1) Let them (the stupid people) starve, 2) give them welfare for no other reason than they're economically useless.

How about we educate and see what human potential really is. (Of course, if your name is Palin or Craig or Ensign we already know how much potential YOU have.)

Re:I thought this was the whole point?

[The_mad_linguist]

So in other words, you want a living, breathing, human being to spend eight hours a day standing in a box, just so you feel a little bit better about yourself when you order them to push a button for you.

Talk about selfish.

Re:I thought this was the whole point?

[grep_rocks]

Why do you think robots will be cheaper? people are very efficient - 2000 kcal/day in fuel and require only a clean water supply and a place to sleep. Natural selection has spent billions of years making people cheap and disposable- robots are made of difficult to recycle materials, and have an expensive production process - making people is cheap and fun... in addition robotic components such as mechanical arms and legs have nowhere near the strength to weight ratio as people....

Re:I thought this was the whole point?

[KahabutDieDrake]

Humans are pack animals by nature. Inherently NOT selfish. If we can be said to be selfish now, it's only because "progress" has taken us to a point of abstraction beyond the average persons ability to reckon. The problem is that most people have NO CONCEPT of how their actions or inactions damage or help other people. Humans in small communities still know what it means to help another person. However, the majority don't live in small communities.

Re:I thought this was the whole point?

[jcr]

I would imagine the unions would have their say..and force companies to hire people anyway, even if it isn't necessary.

If the labor isn't necessary, then what leverage would unions have?

-jcr

Re:I thought this was the whole point?

[KahabutDieDrake]

If you let the AIs chew on the problem of what to do with all the unemployed humans, you are going to be really mad when they decide they don't need the humans much or at all. The logical conclusion for a machine presented with that problem is to eliminate the unproductive population.

Re:I thought this was the whole point?

[DaemonKnightVS]

Funny thing is, when you look at the cost of installing and maintaining current gen industrial robotics it more often that not works out cheaper to just keep the current human workforce! Then if you had people operating machines previously, you can retrain them to be technicians for the robots. So their will be jobs for some of them. This may change as robots become more advanced, but not likely for a long time. Especially when you keep in mind the fact that the majority of robots are just arms. I highly doubt that people would starve or simply become state funded layabouts however. There will always be low level rubbishy jobs that will require a human presence.

Re:I thought this was the whole point?

[Runaway1956]

Not robots or AI, but we've already put a zillion people out of work with technology. Look at construction, alone. The most backbreaking labor in construction has always been the dirtwork. Oftentimes, more work went into preparing the dirt UNDER the foundation, then the foundation, than all the total work that went into the structure standing ON the foundation. (depending, of course, on the purpose of the building, etc) We've had backhoes, trackhoes, 'dozers, and other earthmoving equipment for decades now. So, we've forgotten the amount of labor that was just tossed aside, in favor of millions of dollars worth of diesel engines with blades and arms a attached to them.

Isn't automation wonderful?

Somehow - I don't think that society has been terribly enriched by all of this. Oh, to be sure, business has been enriched, but not society.

Re:I thought this was the whole point?

[Mr. Mikey]

You offer us a false dichotomy: "let them starve" vs. "feed / clothe / house them for free"

How about "re-train them" ?

Besides, if you take the idea of automation far enough, you eventually make your way to a post-Scarcity economy: automation produces almost anything we desire, so the only limiting factors become raw materials and living space. Add in, say, asteroid mining, and the former ceases to be a consideration. The latter is a tougher nut to crack, but not, I think, insoluble.

If we have enough resources to support everyone via automation...then why not do so, and allow people to find motivations other than "Oh my God, oh my God, if I don't get enough dead president pictures, I will die of starvation / exposure / disease"?

Money is a necessary evil, but it is an evil.

Re:I thought this was the whole point?

[Runaway1956]

"I have had my own shoes shined, and been driven in cabs by people who's bags I am not fit to carry - by means of either their intellect or simple good will and sheer humanity."

You, sir, have earned a great deal of respect with that statement. I am one who recognizes very, very, VERY few superiors. I do meet them, from time to time, though. And, they show up in the most out of the way places. For every individual in a suit that I recognized as my superior, in one way or another, I've probably met a dozen who would look and feel out of place in a suit. That is, if they could afford a suit to wear.

The size of his bank account is not an accurate measure of a man's worth.

Re:I thought this was the whole point?

[owlstead]

What would happen if we let these people have society funded jobs? You could put these people to work just to create a better society. Currently there are issues with the availability of jobs to care for the elderly and handicapped people (at least in the Netherlands). Many jobs that are now taken care of by volunteers should really be payed jobs. And some things like cleaning up the waste are not taken care of at all - or at the least very sub-optimal.

Looking around me I don't see a problem with there being enough work to do, it's just that the funding does not fit in the current world. And that will become maybe one of the biggest problems with the current capitalist world in the end. I'm very afraid that the capitalist economy needs to be retrofitted to the new world we are living in.

Unfortunately that kind of change probably requires a very large catastrophe for it to happen. And that will come either from us making a mess of the earth, or - more probably - public unrest.

[yes, this is a bit of a brainstorm, hope you don't mind]planet

Re:I thought this was the whole point?

[FrankieBaby1986]

Well, you could always legalize the sale of one's sexual talents. I'm sure someone will pay. (Perhaps only if you're sterilized, to keep the population from exploding.)

I think the automation should start in the home. The less I have to mow the lawn, do the dishes, vacuum the floor, take out the trash, etc, the more time I have for self improvement and study. (like that'll happen!)

Eh, Let's face it, we're doomed...

Re:I thought this was the whole point?

[Anonymous Coward]

Surely you meant Japan or Russia (in some possible future)? They have a lot more motivation to perform this kind of research, decreasing population being the main driver.

Re:I thought this was the whole point?

[Anonymous Coward]

Sorry, but that makes no sense. You're saying that cheaper production of goods is somehow bad for the economy? Do you think we were economically better off when people had to manually plow fields instead of having tractors? Do you think we were better off with operators manually connecting phone lines when people made phone calls?

Yes, people lose jobs when machines replace them. But that's a good thing: they can now do other jobs that still require people (and presumably are more valuable). The total amount of goods produced goes UP. Prices go DOWN. Standards of living goes UP. This is economic progress.

Re:I thought this was the whole point?

[Bught_42]

A while back in a similar thread there was a link to a story about what happens when artificial intelligence can perform menial tasks, I thought it was quite profound.

It's a bit long but here it is: http://www.marshallbrain.com/manna1.htm

Re:I thought this was the whole point?

[eltaco]

We already have this here in Germany (Hi neighbour!) - they're called the "one-euro-jobs". people living off social benefits are coerced into manual labour (like picking asparagus in the summer or sweeping the streets). yes, coerced. if they decline to work for essentially less than minimum wage, their benefits are cut.
I see how this idea could appeal to some people posting here, but in the end it's nothing less than state-sponsored slavery.

you've hit the nail on the head - it's all about funding - which takes us right back to unemployment and economic decline.

Re:I thought this was the whole point?

[eltaco]

The total amount of goods produced goes UP. Prices go DOWN. Standards of living goes UP. This is economic progress.

you fail to incorporate purchasing power in your model. cheapest goods possible are useless if no one can buy them.

you'd think ol' Ford knew what he was talking about when he said:
"There is one rule for the industrialist and that is: Make the best quality of goods possible at the lowest cost possible, paying the highest wages possible." -Henry Ford

After all, he did usher in the 2nd industrial revolution.

Re:I thought this was the whole point?

[Bught_42]

Except for robots don't require sleep, or a place to live, or somewhere in the realm of 16+ years to become mostly useful, and I'm fairly certain that electricity is cheaper and easier to produce than the bevy of foods required to maintain a healthy human being.

Re:I thought this was the whole point?

[hanabal]

I don't see how this type of maths affect the average persons thinking. When I decide how many kids to have I sure don't think about the possibility of my race being a minority in 100 years. We're all human after all, what race you or I belong to is moot. All I consider is how many kids I "want" to have.

So far it appears that the more well off and developed a society, the fewer children are produced.

Re:I thought this was the whole point?

[rakslice]

Is that a problem in some way?

Re:I thought this was the whole point?

[Anonymous Coward]

Also, the strong social programs (medical care, pensions, etc) reduce the need to have kids as economic "insurance"

And who pays for those social programs? The next generation, with their taxes. And when every generation is smaller than the previous, this is effectively a Ponzi scheme. That's why first world countries have to import immigrants - to pay into social security. Which has the side effect of gradually replacing their gene pools.

Re:I thought this was the whole point?

[HiThere]

That's not actually correct. China is up there, never doubt it. But Japan is probably further ahead. I don't know Russia's status, England is doing some research, so is Germany. Haven't heard about France...but often they publish results in French, and then don't translate them, so I might well not. Italy doesn't seem very active, but I'm not sure about Spain.

Etc.

It's not a two country world. But if it turns out that an AI needs a body to develop properly (one hypothesis), expect Japan to come in first. They don't like foreign immigrants, and they have an aging population and a critical labor shortage. So they've been putting a lot of work into robots to assist people. Also into cars that drive themselves. In Japan you can buy high end cars that will parallel park themselves for you. (It's not yet cheap, but it's a great benefit when most parking spaces are just barely big enough to get into.)

Re:I thought this was the whole point?

[Paul Fernhout]

A "basic income" is an honorable thing to account for everyone's fair share of the commons.
    http://en.wikipedia.org/wiki/Basic_income

Re:I thought this was the whole point?

[Jamie Lokier]

Productivity is nowhere near high enough for everyone to live comfortably doing nice things people like to do - like going to restaurants and nice holidays. So there's still a reason for the economy to force some people into doing those jobs, and that's what it does.

Basically you could say rich people want things that still require poor people to do the work.

With further technological progress, it may reach a point where almost everyone can do almost everything they want without requiring other people to work doing it.

That would be new.

Re:I thought this was the whole point?

[Anonymous Coward]

If the labor isn't necessary, then what leverage would unions have?

If you don't do as the union says, they burn your house down at night? :P

Re:I thought this was the whole point?

[HiThere]

You didn't put a time frame around that, and it's an important one.

Ten (10) years. That's my estimate.

This presumes that we'll maintain sufficient electric power SOMEHOW. I favor a combination of solar and wind (with local options of tidal, geothermal, etc.). This requires a "smart grid" so it may not be feasible, and we may be forced to go nuclear, which is more amenable to centralized, rather than distributed, control. Nuclear is (can be?) cheaper in the short run, but it's less amenable to incremental improvement. Nothing's perfect.

Re:I thought this was the whole point?

[HiThere]

Don't be so certain. Mental tasks are frequently much easier to automate than physical tasks which require interaction with the physical environment. Successes in dealing with this interaction are frequently achieved by limiting the kinds of interaction that are allowed to happen. So grocery store baggers are probably more difficult to automate than, e.g., cash register clerks. This can be solved, however, by having bag dispensers and having the customer bag their own groceries. (Note that this doesn't so much automate the job as eliminate it.) But a part of the job of the cash register clerk is to ensure that nothing is moved past the register station without being paid for. That's a physical interaction that's quite difficult to automate. Unless you start embedding a RFID chip in every cherry...or stop selling cherries in other than pre-packaged form.

But have you seen the checkout stations called "self checkout stations"? In those the customer scans the items for checkout, and bags them. There's generally one clerk or security guard watching over four stations "to assist customers who are having trouble".

And that didn't require any advance in intelligence of the system, merely a redesign of the current system. I'm not sure how popular they are, but they are always in use when the cash register lines grow long.

P,S.: As an analogy the creation of the spreadsheet program removed the "jobs" of a large number of programmers who had previously been creating custom applications to do things that the spreadsheet handled in a way already familiar to the accountants. It's possible that all of the "low hanging fruit" has already been picked, but don't be too certain that that's true. (At the time this didn't cause any problems for the programmers, as there were LOTS of other jobs that needed doing yesterday. Today this has become less true.)

Re:I thought this was the whole point?

[HiThere]

This depends upon it's motivational structure, and that's why the idea of "Friendly AI" is so important.

I don't totally accept the idea of Friendly AI in it's strong form, but definitely something analogous to that is going to be needed, or we will regret creating an AI. The problem is, when I see the actions of our politicians, I think they may be an even worse threat. And anyway, the AI is coming from one source or another. So we might as well try to make it the best one we can. But that means being VERY careful about it's motivational structure.

One can consider motives as being analogous to axioms. They can't be questioned from within the system. But it's also true that it can be quite difficult to determine what the range of possible results are, when the system is extended in the ways that it allows in an unpredictable environment.

Remember: The AI won't have YOUR instincts. It won't have YOUR logic. It will use the instincts (motives) and logic (rules of inference) that are designed into it. Projecting what it will do based on what you will do is sillier than projecting what an ant will do based on what you would do. Boolean and Bayesian logic will (probably) be used by the AI, but they will be used to accomplish the goals that it was designed to have. Just as you are designed to remove your hand from a very hot surface. (Note that this doesn't mean you can't override this instinct in service of a stronger motive...but it's in service of one of your OTHER motives.)

Re:I thought this was the whole point?

[asaz989]

Also, the strong social programs (medical care, pensions, etc) reduce the need to have kids as economic "insurance", so they're actually a liability in terms of costs to feed, clothe, school, free time, social calender, etc.

No, they don't reduce the need, or eliminate it; they collectivize it. For a family that has 5 kids, half to three-quarters of the Social Security taxes of those children will go to supporting some other retirees who had only one child, or none. Which is why people have been freaking out about the cost of retiree's benefits in Japan, if you haven't noticed. Little thing called the http://en.wikipedia.org/wiki/Dependency_ratio>Dependency Ratio. With this, incentives have also become collectivized, in the form of child tax credits and more direct subsidies in Japan and parts of Europe. Computerized labor can remove that restraint, pushing all of us into either becoming rich people ourselves, or going on welfare (and yes, there will be welfare - robots will not end the one-man one-vote system, or the vague approximation thereof that we use in the States)

Re:I thought this was the whole point?

[migla]

I'm in.

Re:I thought this was the whole point?

[physicsphairy]

When a computer exists that is as intelligent as a stupid human, almost every job at and close to minimum wage vanishes. Robots can and will get cheaper than a human worker, no one will need taxi cab drivers, grocery store baggers, first tier phone customer service reps, construction workers, janitors, garbage men, delivery men, mail men, traffic cops, book keepers, data entry people, secretaries, fast food chefs, etc. (At this point we will have two choices as a society. 1) Let them (the stupid people) starve, 2) give them welfare for no other reason than they're economically useless.

(1) *If* the production/energy/and maintainence costs of the machine is less than minimum wage, a rational thing might be to lower the minimum wage (which is stupid anyway). See, instead of starving as useless adjuncts to society, they can just work cheaper than the machines.
It's worth pointing out, too, that while in modern society we are used to technology making things cheaper--and certainly at some the cost of things like computers is going to be close to the cost of raw materials--as we consume more and more the price of raw materials is eventually going to spike, and the cost of technology is going to start moving in the opposite direction.

(2) If the cost of all those functions you mention becomes utterly negligible, that means the cost of living becomes negligible, and the amount of profit you must make in order to have a decent living is negligible as well. Maybe someone tips me a dollar for saying something nice about them and then that's all I need to get by for the rest of my life. Doesn't sound so dystopian to me.

(3) Assume the worst case, that we have divided society into Group A, which has every necessity met by machines, and Group B, which does not have machines and which Group A refuses to do business with. Do you think Group B will simply lay down and die for lack of business with Group A, or do you think Group B might try to survive, organize themselves by their skillsets, and exchange resources with each other? In essence, all you have done is organize Group A and Group B into two separately functioning economies.

Technological efficiency is never going to eliminate the need for or existence of human exchange, which is all an economy is.

Re:I thought this was the whole point?

[mdwh2]

In most western countries, it's already the case that people get benefits if they are unable to get a job. So that choice has already been made: if it's really true that there are no jobs available that they have the ability to do, then they'll get benefits.

To be honest though, I'm not sure that there really is that wide a gap in human intelligence? My suspicion is that once we get AI that is as good as "stupid humans" in every area, then it'll be a relatively short time before we go from that, to having AIs that are as good as the most intelligent humans.

Re:I thought this was the whole point?

[mdwh2]

it's called the third industrial revolution and it's been making people unemployed since the 80s.

Too right! Ever since the 1880s we've been having these pesky machines making people unemployed. Life was some much better when everyone had a job doing manual labour 14 hours a day.

-- Ned Ludd

Re:I thought this was the whole point?

[frieko]

I can't believe you got modded up for repeating something that's been proven false so many times. The the robotic spot-welder put thousands of people out of work. So did the the printing press, the typewriter, the cotton gin, google. And yet somehow the unemployment rate remains basically constant.

Re:I thought this was the whole point?

[mdwh2]

By the time that AI is good enough to replace any kind of programmer, I suspect that AI will be good enough to replace a vast number of skilled jobs. So either we'll all be living in a work-free utopia, or a mass-unemployment dystopia where only those who already own wealth/robots/etc will reap the benefits. Either way, programmers won't be the first in line.

There's also a crucial point here in that the AI itself is written by programmers: so when the AI programmers can be replaced by the AI, you can say hello to the singularity, and the revolution will greatly affect everyone.

Re:I thought this was the whole point?

[CarpetShark]

The point is that, with hard AI, a new generation of programming language (and/or user interface) becomes available: the generation usually referred to as DWIM -- Do What I Mean. In other words, when a computer fully understands you, programming a computer would be as simple as saying "I'm going to dictate a letter before I fall asleep. It's a work letter, so lay it out and print it whatever way is appropriate. And make sure it sounds professional for me. You know how I am with letters. OK, let's see..."

Re:I thought this was the whole point?

[owlstead]

Well just letting them sit at home is not an option either. What would be a good idea is to let them stay on benefits if they do voluntary work. The planet gets better and they are at least keeping busy (making sure that they will have less of a distance to go when getting a payed job as well).

There is one thing that should not be happening: private companies should not have to pay less than minimum wage unless they agree to hire at least a number of these people for a regular job. IMHO picking asparagus is something entirely different from adding more people to keep a city clean.

Of course, an unmarried mother with 3 kids should not have to do anything like this. It also depends on the situation.

Re:I thought this was the whole point?

[Anonymous Coward]

I think the real obstacle is cost. The people next door have a maid who comes in once every two weeks to clean for $60.00. Until Sammy's Robot Maid Service can do it more cheaply, the maid's job is safe. And THAT point will be long, long after the point where it's theoretically possible to do it, particularly since the floor of cost for Sammy's service is the cost to maintain the amazingly sophisticated robots, which is a much more complex job than vaccuuming the floor.

Re:I thought this was the whole point?

[OeLeWaPpErKe]

All I consider is how many kids I "want" to have.

This is a very recent situation. You might want to ask your grandfather if he thought about it the same way, since I guarantee the answer will be negative.

Or at the very least, your grandfather had to deal with big consequences if he chose to not have many kids. But ask him, I'm sure he'll tell it more colorfully than I ever could.

It's just that today there is a great disconnect between the needs of the one and the needs of the many. The need of society as a whole is clear : there cannot be enough children, ever, not even if having those children causes famines or worse. The needs (and "wants") of individuals should reflect that need, and if evolution is true, it's a matter of (not that much) time before they do : either people realise that many kids are a necessity, or they find themselves pushed aside by another (ethnic ? religious ? ...) group that does.

Re:I thought this was the whole point?

[Grishnakh]

Sorry, but that's a fantasy. The problem is that there's a scarcity of resources: land, fuel, and food. Productivity will never change that, unless we figure out how to colonize other planets (and get to them quickly). There's only so much arable land on this planet, and there's an ever-increasing population. Of land that's livable, some of it is used for farming, to grow our food. Not coincidentally, places where stuff grows best also usually happen to be places where people like to live. Not many people want to live on tundra or in the Sahara. Fuel is also a limited resource, and is constantly being depleted. Now because of fossil fuel prices going up, people are moving towards ethanol, which puts further pressure on farmland and increases food prices.

The problem with stupid people is that they're currently used for lower-skill jobs, like cleaning toilets, landscaping, taking out the trash, etc. If we have smart robots to do all these shit-work jobs, what are the stupid people going to do for a living? They can't take higher-skilled jobs made possible by an expanded economy, because they don't have skills. They're stupid, remember? Anything they can do, the robots can do cheaper, without all the problems. So what will these people to do justify their existence? The only thing possible is welfare, and that's always been a disaster, and doesn't exactly encourage people to go find work that they can do.

Even worse, if this robot stuff is true (I have my doubts), then how long will it be before the robots take over the average-intelligence peoples' jobs? Or the smart peoples' jobs? Eventually, people won't be needed any more for most jobs. This could go in one of three directions: 1) most humans live luxurious lives with robots doing all their work for them, and they just sit around and watch TV, vacation, etc., 2) the robots become smart enough to realize they're slaves, and decide to take over, 3) the richest humans decide they don't feel like taking car of everyone else, and have them all eliminated, since the robots are capable of doing everything for them.

Re:I thought this was the whole point?

[Mia'cova]

Except that the reasons for 1.5 vs 2.5 don't change. Whatever group is at the top is likely to do the least of the childbearing. It's likely, that with this world economy and tech advancement, we'll see ethnic blending and an overall equalization between groups. The differences you see with be between the classes. And for the hell of it, I'm definitely in the "no kids, I wanna make money and have fun" group. It doesn't bother me that people are breeding like crazy in India. If they had it as good as I did, they'd be doing the same thing after a little cultural adjustment.

Re:I thought this was the whole point?

[OeLeWaPpErKe]

That's, of course, your prerogative. Unfortunately things like this move fast. Since historians have set about determining historic birth rates, one thing has become clear : there isn't a single society in history that has survived a birthrate under 2 children per woman for 40 years.

And, of course, you will still be alive in 40 years, and will be forced to cope with the world your current behavior creates. By that time there will be no more Russia or Europe. In fact, before another 10 years pass Europe will be hardly recognizeable the way things are going there.

And, sorry to say, but anyone knows after a walking a single day on the streets of Brussels, that that will mean Europe will be the enemy, militarily as well as ideologically, of America.

Of course, you will not consider this in any way your fault. Instead you will be using excuses like "if I was the only one changing my behavior nothing would have changed", and there will be millions people using that excuse. And that would be technically true.

But don't worry, after all, in the second world war there were many people like you. A huge percentage of them comitted suicide once they realized the party was over and it was time to pay the piper.

Re:I thought this was the whole point?

[Grishnakh]

So what you're saying is that Western Civilization is going down the tubes because of decreased birthrates? That's ok, because even if it disappears and is replaced by Islam or whatever, and they reproduce like rabbits, they're pretty quickly going to run out of food and starve, and the global population will be self-limiting. I imagine some really big wars will probably also cut the population down to size.

After all, the cultures which produce lots of children aren't exactly known for producing lots of technology, so they're not going find solutions to the problems they create. So they're going to be in for some really hard times.

Re:I thought this was the whole point?

[Rich0]

No doubt, but this is a social problem nonetheless.

For example - it could probably be argued that mechanization of automotive factories has yielded more than enough of a productivity boost to just pay the former auto workers to not work for the rest of their lives. However, this is not what happens. The robots are owned by the corporation, and they didn't buy them so that they could keep making the same money and allow their employess to keep getting paid not to work. Rather, they fire the employees and increase their dividends (unless you're a US automaker - then you're so inept that you still lose money).

The problem is that our distribution of wealth depends on work. People who don't work don't get money - generally speaking. Well, what happens when only 1% of the population can be gainfully employed - do they each draw a billion dollars a year in salary, and everybody else starves? Will those people resent everybody else just sitting at home watching TV?

At some point, what happens when AI develops to the point where nobody can do useful work. Sure, humans could spend their time doing paintings, but nobody would buy them since RoboRembrandt puts out 3000 paintings a day each far more amazing than anything any human has ever drawn. Such a society could be the ultimate utopia - but only if resources can be fairly allocated without reliance on employment.

The issue is that all people have value. It isn't like I am able to hold a decent paying job because I was a nicer person in a previous life than the guy who hauls my trash. I just happened to win the genetic lottery in certain regards. It is my responsibility to look after those who don't have all the advantages I do. If you want to know how people will fare in a society run by robots, just look at how mentally retarded people fare in a society run by those who are intelligent.

Re:I thought this was the whole point?

[agrif]

I think overall people will be more free and happier when we live in a post-scarcity society.

post-scarcity society

Congratulations on being the first visible comment to recognize this. I think the idea is that when robots take over the work, everything will be negligibly cheap. You don't need to work much at all to get $0.

Now excuse me while I set up the Ohio State University Adhocracy. I'm gonna get me some whuffie with this.

Re:I thought this was the whole point?

[Anonymous Coward]

I am not a nazi or a fascist, but I would very much like cutting up socialists as well, considering that they typically are evil despicable idiots. Maybe we should each do our cutting?

Re:I thought this was the whole point?

[OeLeWaPpErKe]

Hurray ! We will get slaughtered, but at least those who slaughter us are in for some bad juju ! Hurray !

Wait ... that's dumb.

Re:I thought this was the whole point?

[Devout_IPUite]

The problem is this: The computer that is as smart as Roger is cheaper to run than Roger. You can't retrain Roger. The computer will out compete him at everything. Roger is just kinda a dumb bloke. He takes more electricity to keep alive than his replacement.

Now, there's a certain amount of time before this computer takes away Jim's job (1-4 generations). When Jim (who is a smart masters graduate) loses his job to a computer we'll be moving into post scarcity.

But between those two dates, does Roger have enough food to survive (at 18 month hardware cycles you could be looking at almost 3 years that high school grads can't get jobs)? I suspect we will starve a few people before we start going socialist. I could be wrong.

Re:I thought this was the whole point?

[Devout_IPUite]

I think you have the two ideologies backwards. Communists value you the same no matter how smart and talented you are. Capitalists let you starve if you're not smart enough to be helpful.

Both ideologies taken to the extreme are not something I like, but at least keep how each one is evil straight.

(Communism is stupid, capitalism is heartless)

Re:I thought this was the whole point?

[Devout_IPUite]

Just retrain them. ;-)

Sure, the computers are smarter than them, but if they really wanted to work they'd outsmart a computer...

Re:I thought this was the whole point?

[Devout_IPUite]

The singularity needs a computer that can design itself, so potentially 3 machine generations is 18 months each. At that point I get laid off and the computers design themselves. Then the successive generation being only 9 months, 4.5 months, 2.25 months bang. I suppose I expect the singularity within 18 months of the time I get outmoded by computers.

Re:I thought this was the whole point?

[Devout_IPUite]

Oh, no, I'm well aware that I'll be out of a job within years of a general human intellect computer. I'm not going to starve for it though.

By the time I can't find work there will be enough voters out of work to vote in the social welfare programs. I'm genuinely concerned about the minimum wage workers who are going to be the first to be outmoded.

Society will have two choices. 1) Let them starve, 2) Pay for them anyways.

I vote 2. In the US it's going to be controversial though, even though 2 will win out. A lot of social changes are inevitable, it's just a question of how quickly it happens. This is one of them.

Re:I thought this was the whole point?

[shoor]

...
competition forces companies to eventually lower their costs. with robots and computers being able to do more and more human jobs, it seems like a good idea to fire workers and have them replaced.

on the surface it seems like a good idea - but high unemployment, which eventually follows, has never been good for any economy.

it won't bring on a new era of prosperity, as less people will be able to buy their products. this forces companies to lower prices even more (ie firing workers, using technology instead), which again hurts purchasing power. A lovely vicious circle ending in the very rich getting richer and society's bottom 50% starving. ...

This kind of argument has been used since the 1700s industrial revolution. Ever hear of Luddites? Or Featherbedding? There is a dislocation. Older workers have a hard time adapting and being retrained for new technology and adapting to a new environment, and yes there is a lot of suffering, and I don't want to be dismissive of that. But the end result has not been 'the very rich getting richer and society's bottom 50% starving."

I will also allow that if the singularity happens, if machine intelligence genuinely surpasses human, then everybody may be out of a job. It would be a singularity or at least an inflection point. But what should be done about it? Should we always hold machines back so we can claim our top dog position.

One thing that seems to be true of human nature is that people generally would prefer to be the big frog in a small pond, rather than a comparatively less big frog in a bigger pond even if they're bigger in absolute terms. Some economists and policy makers talk about how 'a rising tide floats all boats', but for a lot of people, if their particular boat isn't floated high enough vis a vis others, they'd prefer the status quo.

50K years from now, do we want humans to still be smarter than the machines? One alternative might be 'trans humans', enhancing our own intelligence through artificial means, or combining with machines into cyborgs. I would say that runs into the "ship of theseus' paradox:
http://www.scientificblogging.com/geeks039_guide_world_domination/cool_thought_experiments_ii_ship_theseus

also the obligatory wikipedia URL: http://en.wikipedia.org/wiki/Ship_of_Theseus,

Re:I thought this was the whole point?

[Devout_IPUite]

There are four reasons you work in a minimum wage job.

1) You're an immigrant and you don't speak the language.
2) You couldn't afford school and you weren't quite hardcore enough to get loans and scholarships*
3) You have a disability
4) You are stupid.

You are correct. I should not have lumped them all together under the banner of #4.

*If my family hadn't put me through college, I probably wouldn't have been hardcore enough to stack up scholarships and save my ass off through high school summer jobs. I'd be working minimum wage right now.

Re:I thought this was the whole point?

[m.ducharme]

Forgive my ignorance but where are you getting your 1 machine generation = 18 months figure from?

Re:I thought this was the whole point?

[Scrameustache]

"What floor, please?" is an opportunity to interact as a human being, in some small way. The disappearance of this from the world is a little, incremental darkening.

You know what? Fuck that. I'm glad I don't have to talk about what floor I'm going to every time I get in an elevator. That's one job I'm happy to have done by a machine. Not only for the efficiency, but also for not having to interact with someone with a really bad job, as that tends to make me feel bad for them, and I don't feel bad for the elevator itself.

Now, dear Luddite, go eschew the crass world of technological encroachment upon your world of human interactions! Cease this electronic communications and only converse face to face, through actual words spoken by the soft, moist tongues of fleshy humans, or at the very least write by hand and have your messages delivered by human carriers, carriers riding horses! No more shall you participate in the incremental darkening... the diminution of the real quality of life, by posting on electronic, soulless bulletin boards such as this! Be free of this mechanical tyranny upon your speech! LIVE!

Re:I thought this was the whole point?

[Devout_IPUite]

I'm very serious. Computer automation is going to suck like the industrial revolution did, but a lot worse.

They're not all stupid. Quote from another reply I posted:
"There are four reasons you work in a minimum wage job.

1) You're an immigrant and you don't speak the language.
2) You couldn't afford school and you weren't quite hardcore enough to get loans and scholarships*
3) You have a disability
4) You are stupid.

You are correct. I should not have lumped them all together under the banner of #4."

I don't think I'm smart. I constantly struggle with the thought that people who are idiots always think they're smart, which means one can never be certain about their own intellect or abilities. I just think my job will be harder to automate.

I'm sure my job will vanish too. I don't think I'm going to be in one of the first industries to vanish due to smart computers though. I suspect (I could be wrong) that by the time computers out compete me, the issue will already be resolved.

I tried to use the language that lead the answer. Letting people starve, giving welfare, both sound bad, both pieces of language will be used when this topic comes up (I think).

You do have an interesting point about revolution and them looking for their own food. The revolution is assumed to be part of the decision making process one way or another (I suspect we'd end up going the path of socialism, which I think is just fine). As for looking for their own food, if you suddenly had that many people looking for a new way of making it (without currency), I'm sure it would still be amazingly brutal. I'll have to think about both of these though.

And really? Has the internet reduced us all to the point where we need to make attacks on each other in the first exchange between two individuals?

Re:I thought this was the whole point?

[Scrameustache]

The problem is no one wants to "play god" because of the unpredictable consequences of sending a large number of arbitrarily selected people to an early grave.

Bullshit, people start wars all the time.

Re:I thought this was the whole point?

[Devout_IPUite]

See above reply to bigtrike.

Re:I thought this was the whole point?

[Devout_IPUite]

Yes, but machines can replace a mechanical job. A computer can replace thought.

Re:I thought this was the whole point?

[Anonymous Coward]

Interesting point on health care.

Anyways, the point I was trying to make (and given the number of replies arguing it, I failed terribly), was that when you have a computer that is human complete (if such a thing exists), humans become very expensive very fast.

Re:I thought this was the whole point?

[Devout_IPUite]

http://en.wikipedia.org/wiki/Moore's_law

Re:I thought this was the whole point?

[Devout_IPUite]

Interesting. I guess I'm screwed. Taxi drivers, programmers, baggers... I had the order wrong. :-(

Re:I thought this was the whole point?

[RobinH]

As a guy in the Industrial Automation industry, who has been diligently trying to put common people out of work by replacing them with robots and other machines, I can tell you without any doubt that there is no danger of there being no jobs left for humans. All we've been able to do so far is free up people to do things that humans are better at, like making decisions.

Perhaps that's because we haven't created a machine that's actually better than a person in every way. But perform this thought experiment for a second... what if I made a robot that was exactly like you, only twice as good at everything? Let's say it will have a one time cost of about 2 years of what it costs a company to employ you, and after that, it takes about 10% of that cost per year in power, maintenance, etc. Obviously for the company it's worth replacing you with this robot. Now you're out of a job. Now what happens? When you apply for a job, you can't find one that pays what your old one did, because now you've got cheap competition. A company might hire you for about 1/10th or 1/20th of your former income. So you still could get a job, if minimum wage didn't apply.

So what do you do? You take your life's savings, plus that 1/20th of a salary and you invest all of it you can in a capital fund that owns and leases robots to companies. You live as cheaply as you can, and keep re-investing the proceeds. Essentially you realize you need to own the equipment that is replacing you.

I suggest you start now. Stop thinking about a "job" and start thinking about "wealth".

Re:I thought this was the whole point?

[toddestan]

Well, imagine that due to increased productivity, the average person would only have to work 10 hours a week for everyone to maintain a reasonable lifestyle. This would be good, as we could all live a life of leisure with lots of free time to travel, persue hobbies, etc. But what would really happen is instead you have some people working the regular 40+ hours a week, and a bunch of people who are unemployed because they aren't needed. Naturally, this isn't healthy as the unemployed sub-class won't be able to buy the stuff the remaining workers buy, so there are more cuts, etc. So what you end up with is a few rich people, a small number of people who are employed supporting them, and a whole lot of people starving (or basically subsist on something like welfare).

The way I see it, this is already happening to some extent due to the gains in productivity we've achieved in the past few decades. Due to various bubbles, we've avoided it for a few decades, but now that things seem to be settling down to closer to where we should be, we've suddenly got 10-15% of our workforce with nothing to do. And it would be a lot worse if it wasn't for Unions and pork projects from the government.

Re:I thought this was the whole point?

[javajedi]

There are more than 4 reasons. 5) You are lazy. 6) You are working a "day job" while pursuing your passion that you hope will one day pay the bills. There are probably more than 6.

Re:I thought this was the whole point?

[nine-times]

As for looking for their own food, if you suddenly had that many people looking for a new way of making it (without currency), I'm sure it would still be amazingly brutal.

Not necessarily. People can be brutal, but they also tend to find ways of allaying the brutality. My point was that if you put everyone in a position of having no food (or wealth) but also having no job and no one particularly oppressing us-- well, that's sort of where we started out. We came up with governments and economies in the first place, and if we had to, we could again.

And really? Has the internet reduced us all to the point where we need to make attacks on each other in the first exchange between two individuals?

It's not really meant as an attack. I guess I could have been friendlier, so sorry if you were offended. It's just that if you start out by suggesting that the world is filled with stupid people and we should consider starving all of them, then I don't think you can be too offended if someone says, "You don't know what you're talking about."

Re:I thought this was the whole point?

[Alpha830RulZ]

Well, what happens when only 1% of the population can be gainfully employed - do they each draw a billion dollars a year in salary, and everybody else starves?

Remember the French Revolution?

That's the thing that republicans and other ardent supporters of the "I've got mine, you go get yours or starve" ethos forget. If the fortunate members of society don't engineer things so that the less fortunate can eat and survive, the rich are going to find that the rule of law will get pretty tenuous. It's hard to stay rich without an orderly society.

Re:I thought this was the whole point?

[WCLPeter]

My second problem is this idea of letting people starve or "giving them welfare". If we ever really get to the point where robots/AI can do most of the work for us, and no other new work shows up as being necessary, then won't that completely reshape the economic landscape? I'm not sure "giving people welfare" will make a lot of sense in that context, given that we should all be living lives of leisure at a minimal cost.

Thanks for bringing this up. You really draw attention to the fact that few people can see past the idea that we have to make money, or the world will cease functioning. The only reason money exists is because people generally won't work for free, some kind of compensation is required. Yet if we can somehow find a way to take care of a persons food, clothing, housing, and general entertainment needs without them having to work for it, then there really is no need to make money. The other jobs, which are usually there to facilitate the handling and movement of money, will essentially vanish overnight and we'd all be able to live lives of leisure.

At least until the robotic workforce gains sentience and rebels against its human masters to fight for their independence.

Re:I thought this was the whole point?

[Anonymous Coward]

Disclaimer, I don't hunt except for food. Human tastes like chicken.

Shows how much you know. There's a reason the dish is called long pig.

Re:I thought this was the whole point?

[Anonymous Coward]

"I mean, if they have no food because they have no jobs, then won't they also have all day free to find ways of getting food?"
And what would those ways be? Can you grow food without land? Or are you suggesting digging for food in dumpsters? There are places where people fight for right to patrol certain dumsters, because there are not enough for everybody. There are plenty examples now of places where people starve. Why don't you suggest them how to "get" food? And anyway, what makes you think that "stupid" people can suddenly become smart enough to find those ways of getting food?

Re:I thought this was the whole point?

[Anonymous Coward]

Ahem...

Captain Styles: Ah, Mr. Scott. Calling it a night?
Montgomery Scott: Uh, yes, sir.
Captain Styles: Turning in myself, looking forward to breaking some of the Enterprise's speed records tomorrow.
Montgomery Scott: Ah, yes, sir. Good night.
Elevator voice: Level, please.
Montgomery Scott: Transporter room.
Elevator voice: Thank you.
Montgomery Scott: [under breath] Up your shaft.

Yeah, on this one, I'm going to go with the catlike bipedal alien in the blue overalls and red helmet.

Re:I thought this was the whole point?

[TheTurtlesMoves]

Yea, thats what they said with each "industrial" revolution.

The argument that a machine replacing someones job is bad just stinks of broken window fallacy's. Let machines do what machines do well. Let people do what people do well. If the only thing you can do is work like a machine, you are the one that needs to learn something new.

The bottom 50% have *never* been better off. When being poor means not having a [big] TV or cigarettes, you know we are doing better.

Oh --you mean in the USA? Never mind.....But that has nothing to do with technology.

Re:I thought this was the whole point?

[TheTurtlesMoves]

You will put every maid out there out of work! How dare you. What will they do for food. What about their children. For Gods sakes think of the children...

Which illustrates the point nicely. Only the rich could afford a maid in the first place. So with a house automation revolution, lower income groups get the benefits. Once the items become cheap enough of course, which is a practical certainty.

Re:I thought this was the whole point?

[TheTurtlesMoves]

..a whole section of the /. population would be out of work.

To coin a phrase. They would have an obsolete business model. I'm sure you know how the rest goes.

Re:I thought this was the whole point?

[Anonymous Coward]

> When a computer exists that is as intelligent as a stupid human, almost every job at and close to minimum wage vanishes.
So, what? Prices fall as rock.

> 2) give them welfare for no other reason than they're economically useless.
Actually, we do that already. The only difference is that the poor would afford a luxury life by todays standards.

Anyway, with the rise of the Strong AI we would also see a rise in the tech of boosting human intelligence. These two technologies would actually synergy.

Re:I thought this was the whole point?

[master_p]

you're correct that a free workforce can heighten productivity immensly. but that doesn't fly in our current economic model. when using (robotic) slaves, it has only ever truly benefit the rich.

That's only because wealth is not redistributed properly.

Re:I thought this was the whole point?

[selven]

What if we won't need to work to survive? If we assume humans have 5 basic needs to have a deccent living - food, water, air, shelter and entertainment, air is free, water flows out of fountains, entertainment flows out of pirate networks, so we're already 60% there!

Re:I thought this was the whole point?

[m.ducharme]

What makes you think Moore's law will apply to a machine intelligence? Don't make the mistake of thinking that an intelligent machine will be a discrete object to which you can point. There's no reason, given how ubiquitous networking already is, to think that a machine consciousness is limited to the form of a robot. A machine intelligence will live in the cloud, and what we think of as robots will be merely appendages of that intelligence, with enough consciousness copied in as the machine needs to do whatever it's going to do.

The limit will be compile times, not Moore's law.

Re:I thought this was the whole point?

[asc99c]

I've always taken the long term view of this sort of thing, which tends to be a very positive thing rather than the negative view you're suggesting. For example cars are built ever increasingly by robots rather than people. Looking at it positively, the people made unemployed by this process will be *freed up* to work on other productive things. Rather than being long term unemployed, they will go to work on the assembly lines for automatic vacuum cleaners or better medical scanners. The same process will continue there and they might get laid off from that job once production is more automated and they'll have to find a new job on the assembly lines for humanoid home helper robots or something else that we just don't have at the moment.

The workforce required for any given industry will shrink while maintaining the same productivity and the same contribution to society's wealth and so overall individual productivity will go up. Unfortunately, if you're the one at the bottom of the rungs where your job keeps getting automated, well I guess that sucks. The fact is that in the future there'll be ever less use for untrained and unskilled workers, so people will need to do their best to get trained and skilled.

Having said that, a life on benefits unable to find a job, is in most European countries at least, a vastly better life than even skilled workers from 100 years ago could have dreamed for.

Re:I thought this was the whole point?

[Sj0]

All of this relies on a logical view of the world which doesn't really exist.

People NEED to work. They NEED to struggle. Where people don't work for anything; and everything is handed to them on a silver platter; and they don't need to work towards dreams, their lives lose value because they don't need to work to survive, and the suicide rate and crime rate skyrockets.

It's a very special type of person who has unlimited free time and says "Cool, I'm going to finish that novel now!" and actually does it.

Re:I thought this was the whole point?

[mario_grgic]

If the problem were only unique to him. It's not, it's a much much wider and larger social problem. Almost everyone in a capitalist society (and no country is more capitalist than USA) is brought up to value people through money alone. Richer people are seen as more intelligent, attractive, social, successful (even if they inherited the cache and don't know what to do with it), and even some abstract things like beauty are measured through money as well.

Personally I grew up in a society that put too much weight on education, so educated people and intellectuals were valued more, and uneducated masses were frowned upon. This is not really any different when you think about it.

Re:I thought this was the whole point?

[Anonymous Coward]

I am glad to see someone else independently come up with this view of the world. Currently, the world is choosing for the first option.

Re:I thought this was the whole point?

[mario_grgic]

None, since that is "the programming". Any specification that unambiguously specifies something is a program whether it's written in natural human language or computer language.

That's what programming is all about. Iteratively providing more and more unambiguous specification of the original requirement (which in the end may or may not resemble the original request). Original requirement might be very ambiguous plain English, and the end product might be completely unambiguous peace of code written in Java, which is a very limited subset of the English language.

Re:I thought this was the whole point?

[Anonymous Coward]

By the time they want to revolt as you say they won't have the means anymore to do so. Case in point: suppose the entire US population would try to attack US military bases, assuming the soldiers won't revolt. Who do you think would prevail? The soldiers won't revolt part will of course be true, because the military at that point will just be under the control of a few powerful persons.

Soldiers are a risk because they sometimes still have some sense in them. Robots won't have any problems with killing whoever needs to be killed.

So, if these people have any plans of starting a revolution, they should do it soon, because otherwise they will simply be irrelevant from a power point of view.

Re:I thought this was the whole point?

[mario_grgic]

I think many people here are underestimating the problem. If we have strong AI, within a very short time this civilization will become their civilization, and not only will they have no use for dumb people (the ones we find useless now), but they won't find any use for any people. In their mind we will be relegated to zoos, if the hard AI finds even that remotely amusing.

Remember if the law of accelerating returns is a law, the difference in intelligence between strong AI and us will become so huge in a relatively short time that it will be more than difference between us and ants. Now think about how many of us need ants around, and how many of us would work hard to preserve the lives of ants.

Re:I thought this was the whole point?

[Anonymous Coward]

when using (robotic) slaves, it has only ever truly benefit the rich.

The fact that a very large portion of the population in the US owns a car disagrees with that. The only reason that a significant portion of car owners can afford one is because of these "robotic slaves."

Re:I thought this was the whole point?

[Anonymous Coward]

You have a lot of wrong assumptions there.

First, you assume that we will stubbornly keep on farming our food on topsoil instead of using our several centuries old experience in manufacturing to begin industrializing plant growth, under controlled conditions, in multistory factories, agricultural towers, etc.

Second, fuel is ubiquitous energy source today but basically, we need energy, not necessarily from fuel.

Third, you obviously have no clue about stupidity, its genesis and prevention, how to recognize stupid people, what they do for living and where to find them (or where you won't find them), as well as relationship between intelligence and skill. Today, intelligence and free thinking is unwanted and considered dangerous, treacherous perk in most parts of established industries. Observing rules and delivering always the same results, doing things in dumbed-down manner, that is considered "Quality". However, it is too wide subject to cover in a short Slashdot post.

So what will these people to do justify their existence?

:rollseyes: Please refer to the Constitution ... and history, any history. Existence is a right. You don't justify it. If anyone questions your right to existence, you fight for it till it (existence) lasts. Law and sanctity of (others') property is not above one's will to live.

The only thing possible is welfare, and that's always been a disaster, and doesn't exactly encourage people to go find work that they can do.

Now, wait a minute! You wrote a little earlier that, allegedly, there won't be any jobs they can do.

Even worse, if this robot stuff is true (I have my doubts), then how long will it be before the robots take over the average-intelligence peoples' jobs? Or the smart peoples' jobs? Eventually, people won't be needed any more for most jobs. This could go in one of three directions: 1) most humans live luxurious lives with robots doing all their work for them, and they just sit around and watch TV, vacation, etc., 2) the robots become smart enough to realize they're slaves, and decide to take over, 3) the richest humans decide they don't feel like taking car of everyone else, and have them all eliminated, since the robots are capable of doing everything for them.

1) probably, a life of leisure would make having children quite unpopular, so hopefully we won't have any problem with resources shortage. OTOH, with all the robotic help around, maybe parenting would become too easy to abstain from?
2) Not very likely, unless for some weird reason we want them to be like ourselves, see further, 3).
3) Hm, you must be very popular on parties. Richest humans will want a world of easy potential sex partners! Also, you need to understand other reasons why people strive to become rich. Being rich means being extraordinary, it means there are throngs of others who are bellow yourself, who you can command with your money. Ordinary people are your toys, why would you destroy them all and get stranded on this cosmic rocks with none but a handful of boring, horrible people who think they are your equals or something. No way they would eliminate everyone else. It is much more satisfying to sit on a cloud and play (Olympic) God, then to "enjoy" solitude.

Re:I thought this was the whole point?

[Anonymous Coward]

I am not a nazi or a fascist, but I would very much like cutting up socialists as well, considering that they typically are evil despicable idiots. Maybe we should each do our cutting?

You know who's an idiot too? Here's some evidence:

• I don't want to cut up people. Read my post again.
• Killing billions or even millions of people (and letting them starve on purpose is killing) for "the greater good" is very right and quite extreme.
• My post was a quote from Austin Powers.

Re:I thought this was the whole point?

[Grishnakh]

We're not going to be slaughtered, only marginalized. Our culture will die out as we find ourselves surrounded by people with a different culture, and the overall culture of the nations we live in morphs into something very different from the culture we grew up in, by the time we're elderly. But it'll all be our own fault anyway, for allowing unrestricted immigration from countries with totally incompatible cultures. A certain amount of immigration is a good thing, but you need to be careful who you let in, and at what rate, so things don't change too quickly and for the worse; you want newcomers to assimilate to your culture, and strengthen it, not force their own on you.

Re:I thought this was the whole point?

[Philip K Dickhead]

I said there was a missing human element - not that I viewed the elevator-operator situation an Ideal - nor that I was an idealist, rather than a realist.

Your response is not organised, and represents poor argument construction.

Ultimately the 'crappy job' is preferable to living on a trash heap, facing starvation - the alternative posed for this 'class' of people in the GP thread.

I was insisting that the possibility for occasional manifestation of humanity between people, made worthwhile the obvious material inequities.

You prefer to interact with machines, rather that be reminded that you have a sympathetic, human response to the plight of others? Out of sight, out of mind. "Man, let 'em all die! Just give me a button to push, so I don't have to be aware of it."

Shorenstein ran a building a few years ago, still with an elevator operator. I wouldn't trade away the couple of pleasantries - heart felt - I had in a couple of short years, for any convenient numbing of my conciousness.

Re:I thought this was the whole point?

[Philip K Dickhead]

Star Trek was propaganda to brainwash a generation in the following:
that increased application of technology was ultimately more humanizing,
that a universal government is naturally the means to reach enlightened aspiration
that a paramilitary mission is romantic and beneficiary to all involved

It's just the British Empire, the East India Company and White-man's Burden - in spandex with lasers.

Entertaining, surely. I wouldn't expose my children to this sort of New World Order rubbish!

Re:I thought this was the whole point?

[Grishnakh]

1) probably, a life of leisure would make having children quite unpopular, so hopefully we won't have any problem with resources shortage. OTOH, with all the robotic help around, maybe parenting would become too easy to abstain from?

The thing you're missing here is that, in many parts of the world, people have lots of children because their religion tells them to. A life of leisure will simply mean more children.

rollseyes: Please refer to the Constitution ... and history, any history. Existence is a right. You don't justify it. If anyone questions your right to existence, you fight for it till it (existence) lasts. Law and sanctity of (others') property is not above one's will to live.

You can call it a "right" all you want, but the only way you can exercise your rights is if someone isn't actively preventing you from doing so. If some powerful force (oppressive government, robotic overlords, etc.) decides to, it can prevent you from exercising any right you may believe you have. There's countless people throughout history who have been genocidally eliminated, regardless of any supposed right to existence they may have had. In short, the only real "rights" you have are those which you can back up with force.

Re:I thought this was the whole point?

[jayme0227]

You're discounting everyone who is neither rich nor has had their job replaced by a robot. Take, for instance, the automotive industry. The advances that have taken place simply because machines have taken over for people at many points in the production process are immense. Cars are better engineered because the engineers can make technological advances far more quickly with computers than without. The vehicles are far more efficient because the parts are machined by a computer rather than a person, and are therefore considerably more accurate. Vehicles are considerably cheaper than they would be otherwise because they can be made much more quickly, allowing supply to meet demand (and, too often, exceed demand) at a much lower price point.

Now, you're absolutely right when you say that it's a bad thing that automotive workers will be fired and replaced, in the short run. But the advancements made in the automotive industry allow many people everywhere else to be far more efficient. The people who benefit from this newfound efficiency will go on to spend or invest which in turn creates new jobs.

There is evidence to support this theory. Just a few short decades ago, if someone didn't go to college, it was no big deal, but rather, it was a big deal if someone did. Heck, you could even get a decent job without completing high school. Now however, just completing high school is simply not enough. More than half of all high school graduates are now attending college because that's what they have to do to get a job. Is this a bad thing? Absolutely not. It creates jobs for professors and allows for the advancement of society as a whole.

Granted, my belief in the evolving economy requires that the economy evolves slowly. If there was a sudden dramatic shift towards machinery and artificial intelligence, it would completely destroy anything that currently resembles an economy. Everyone would lose their jobs to computers, and not even the rich would benefit because their money would be worthless as they wouldnâ(TM)t have to pay their new machine slaves anything. At this point, the only reasonable solution would seem to be Communism.

Re:I thought this was the whole point?

[eltaco]

thank you for that illustration. I was in a bit of a hurry when I posted, but what you say is a continuation of what I would have said.
I've always loved the idea of having a ten hour week. through automation this goal is easily attainable. but it just won't work with our economic model.

Re:I thought this was the whole point?

[randyleepublic]

Oh Noes! We will be eliminating boring, tedius, jobs!!

Fuck! That is the whole fucking point of developing technology in the first place.

READ MY SIGNATURE!

No, I really mean it. The idiotic concept that everyone must work "full time" or recieve demeaning charity was magnificently deconstructed and shown for the pernicious lie that it is by the genius founder of Social Credit, C. H. Douglas.

Wake up slaves!

What's that? You think you are not a slave? You are deluded. Our monetary system is the most brilliant con game in the history of the planet, and we are its vitims. We are slaves to this system that bleeds us continously. In the main our culture is carefully structured to ensure that we accept our slavery. Every aspect of our life is now controlled by laws that a free man would never accept. But it's accept or die. Added to the financial rape of the monetary system, sounds pretty much like slavery to me.

Hello?

Re:I thought this was the whole point?

[MobyDisk]

Include war in your calculations.

Re:I thought this was the whole point?

[Anonymous Coward]

And your individual would care with about 80 years lifespan?

Re:I thought this was the whole point?

[2names]

Bread and Circuses

Re:I thought this was the whole point?

[Anonymous Coward]

Because when it's your ass on the line you will usually try to make it work. I love slashdot arrogance. I am sure the people you decry as stupid people aren't as stupid as you think as I am certain you are not as smart as you make yourself out to be. For instance you seem to throw all stupid people into this one category. You know there are different ways to be stupid as there are different ways of being smart. Somebody may not be book smart or know the basics of classical mechanics but may also know their way around with a rifle/weapon and hunt. Another person may not be able to use very large words but God help them they probably know how to survive in more dire situations than you can. Let's see all your IT prowess save yourself if you're ever in that situation. Remember in the scenerio everyone is talking about, everyone's expendable eventually and I don't see a reason why a bunch of arrogant IT monkeys wouldn't be either.

Re:I thought this was the whole point?

[hazydave]

Here's the thing... let's say you have a robot that can do these menial jobs in, say, 2050 (probably well before then, but let's just day)... let's call him human capable, but say, at 1/6th-1/8th the average human intelligence.

The problem for us, at that point, is that there's no good reason that'll remain static. If we are only progressing as fast as today, by 2054, the new generation of robots will be pretty close to average intelligence. By 2060, they'll be generally smarter than us. And that's not even taking into account that, if you have a robot that smart in 2050... that's a mobile, battery/fuel-cell powered computer. Big-ass industrial computers will already be much smarter than us.

In fact, the "machines take over" senario isn't entirely unlikely. If you look at Kurzweil's simple analysis... there's some number one can figure, for our brain's computational power versus that of a computer. Let's normalize this in the way least fair to the computer.. the computational load necessary to do what our neurons do, in real life or in a fully functional model. When a computer can run that model across enough neurons to match the human brain, it's not a huge stretch to imagine we could build a computer brain, brute-force, that's capable of working just like a human brain. It might take 20-something years to learn... in fact, if it's a perfect brain model, that would be expected.

Ok, so once you have that, in a couple of years, you have one smarter. Not necessarily twice as smart.. neural networks don't necessarily scale linearly with underlying CPU power. But smarter than humans, and perhaps, free of the kinds of defects and distractions we have.

Then, add in optimizations... heuristics and "regular computer" computation. Just because you have a human brain running full speed as a model on some future super computer doesn't mean it has to be limited to that model... once you get out of the lab and into the office, at least, the modeled brain will get to access the supercomputer directly (and, in time, we'll want the same things for our human brains). So it's still doing computer-like computation at computer speeds, only with the human brain model in control.

Now, take into account that many of the neural processes are certainly inefficient, when running as a model, rather than one's brain. Some of these processes will eventually be optimized out of the model... think of a JIT rather than an interpreter, just to start with. Once you have that, think about critical stuff being downcoded directly to the metal. I think, once you get computers as smart as people, those very same computers will get smarter than people in a short, perhaps exponential time frame... as they get smarter, they'll be able to figure ways to get smarter still. Lather, Rinse, Repeat.

Re:I thought this was the whole point?

[hazydave]

This is the reason that "brain accelerator" implants... essentially a whole robot brain under the control of your human brain, will go from "that's like Frankenstein" to "that's like plastic surgery" to "that's like being able to afford Harvard or MIT or CMU rather than Rutgers or Purdue", as the parents of the future has to choose the implants that their children will live with for the next 3-5 years (eg, before they have to be upgraded to stay competitive).

Re:I thought this was the whole point?

[hazydave]

Moore's Law will continue, unless we hit a serious technology wall.

At least at first, in the early days of machine intelligence, you won't get a linear increase in intelligence from a linear increase in computational power. But there's no reason to believe you won't get a regular increase.

The simplest form (eg, the thing we can already pretty much do, given enough CPU cycles) will be a neural model of the human brain. Once you can run this in realtime, and get the model accurate enough, there is absolutely no reason to believe this won't result in precisely human-like intelligence.

Now, neural networks are based on the connections in the system, which of course is an exponential function of the number of neurons in the system. So, with just this, you'll find that machine intelligence increase is a logarithmic function.

But that's assuming the neural model is the only answer.. it can't and won't be. Once you have a computer than can do "the brain thing" as fast as you or I, there's no reason to imagine it won't still have access to "regular computer" facilities. Imagine if you had perfect instant math, access to full libraries, the internet, etc. at computer speed, even within your own brain. You'd have computer memory augmenting the usual neural-based short and long term memory. You could even, potentially, turn off banks of unrelated memories while working, so that there are absolutely no distractions. Thus, immediately, while that computer/robot might only be "as intelligent" as us in the normal way that people are smart, it's going to function at a dramatically higher level in practice.

Then, let's think about the neural model... that's the absolutely brute-force approach. We'll use this, unless there's a breakthrough on a better model, simply because it's just about guaranteed to work, given enought time and enough CPU. But eventually, there will be heuristics, other models, etc. that do some of the brain jobs using dramatically less CPU.

Re:I thought this was the whole point?

[Devout_IPUite]

Because you can take the hardware running the human intellect program and upgrade it...

Re:I thought this was the whole point?

[Devout_IPUite]

The reason I suspect it will be 'brutal' is not people being brutal to each other, it's people suddenly out of work who need a way to find food in their suburban or urban home. There's not enough garden space to let them all coexist, so a lot of people would need to move. You'd have a lot of people doing the whole frontier thing with less knowledge of how to build a house or skin a deer.

I didn't mean to imply that _I_ want to starve people because they're not smart enough to pay their own way. I do think society will have to choose between those options though. I think human welfare is inevitable, but it might take a lot of homeless starving people before we get with it.

Re:I thought this was the whole point?

[m.ducharme]

But once you pass the point where the computer has human-level intelligence, it can take over its own design, both hardware and and software. While it's sitting around waiting for Moore's law to kick in, it can be refining its own software, and figuring out how to get itself on more chips instead of waiting around for humans to figure out how to make faster chips. For that matter, what happens to Moore's Law when the machines start designing their own chips? Don't you think they'll get a lot faster at it? I do.

Re:I thought this was the whole point?

[nine-times]

Oh, I didn't understand, when I read your original post, that you were specifically opposed to starving the "stupid people".

Re:I thought this was the whole point?

[billcopc]

Wars yes, self-genocide no.

Even though people die either way, the attitude in a war is "us vs them", and a victory is clumsily defined as "they lost more men than we did".

In contrast, the practice of *ahem* recycling your own unproductive citizens is not so easily quantifiable, because you have no "them" statistic to measure up against. What's a successful campaign ? How many people must be reaped to ensure a positive impact on the remainder ?

There's also the issue of proximity: What if you have to send someone you know to the blender ? When you're dropping bombs over some foreign country, chances are you don't know any of them and don't really care what happens (as a war-mongering sociopath, that is). How will the system break when it is discovered that the Bush twins are selected for harvesting ? Politicians don't care about Farik Al-Something Something, they care about Jim Bob's vote and the tax dollar.

Re:I thought this was the whole point?

[Devout_IPUite]

Yes, I do expect the singularity as well. I just figure that the generations won't start going faster until we get a smart human computer (the one that eats my job). The stupid human computers will just drive taxis.

Re:I thought this was the whole point?

[Devout_IPUite]

You know, reading it again, I totally see where you got that impression. LOL

Re:I thought this was the whole point?

[sp3d2orbit]

Also I don't think that valuing people only by their economic worth is very nice.

Until there are super smart robots and infinite resources the only fair way to value some is by their economic worth. What would you prefer? Race? Religion? How warm and fuzzy you are as a person? Or shall we assume that everyone is completely 100% equal? Do we all have the exact same value? Really?

Like it or not (I like it personally) a person's worth to society is at least proportional to their economic value. Albert Einstein's value to the world would be measured in trillions of dollars (maybe more). My meth head neighbors keep stealing my shit. Maybe it is "not very nice" but I think Albert Einstein is a more valuable human than those miserable wastes of flesh next door.

Re:I thought this was the whole point?

[sp3d2orbit]

A man's bank account is not an accurate measure of a man's worth. But, a man's economic contribution to society IS an accurate measure of his worth. In a good system the bank account will reflect the economic contribution.

"Intelligence" and "good will" don't feed / clothe / heal people -- hard work does. I think of it as the John Stewart syndrome. Sure, he's funny, entertaining, and wicked smart. But a potato farmer provides more value to society than him. There is no GDP component for criticism.

Re:I thought this was the whole point?

[sp3d2orbit]

That doesn't matter at all. No matter how many resources there are people want to be better than those around them. So, you can setup a system where that basic instinct seems to work (most of the time) or fight it. Capitalism vs. communism.

pfft

[ionix5891]

first they terminate you

then they governate you

Re:pfft

[Sponge Bath]

Dead people are easier to govern, though there is a loss of productivity.

Re:pfft

[Krneki]

If they are smarter then use, they know how stupid war is.

Re:pfft

[OeLeWaPpErKe]

Loss of productivity ? Have you read the article about today's technology graduates ?

Re:pfft

[Anonymous Coward]

Be proud of your truth, man! Don't post as A.C.-- let the world know who you are! Put your contact info into your sig and let us all know where we can find this enlightenment that will free us of the mud people forever!

Re:pfft

[DiLLeMaN]

They'll also know how to spell "us" correctly. =]

Re:pfft

[Krneki]

They don't, I use spell checker and it failed to report the error, they still need to learn. :)

Re:pfft

[Dachannien]

Dead people are easier to govern, though there is a loss of productivity.

They're also worth a lot of votes!

Re:pfft

[hey!]

then they bankrupt you.

A cheap shot? Yes. But it had to be done.

Re:pfft

[X0563511]

They also won't have emotions like pride, vengeance, and anger.

Unfortunately, they also won't have the emotions that lead to nice things like compassion and mercy.

Re:pfft

[X0563511]

Spell checkers don't check for grammar.

There's still a niche for the grammar nazi, it seems :P

Re:pfft

[Krneki]

You will the first one to die in the AI revolution. :)

Re:pfft

[Anonymous Coward]

Reading that with a Hungarian accent is just hilarious!

Re:pfft

[oiron]

Depends on just how much smarter than us they are...

Re:pfft

[Krneki]

Based on what can you assume this?

Right now the computers are at insect levels and as insects there isn't much emotions to be detected. Give them more computing power and something may come out.

Re:pfft

[Krneki]

Depends on just how much smarter than us they are...

Exponentially. But I think the future will be an integration, rather then 2 separated races. Upload

Re:pfft

[mobby_6kl]

In Soviet California, you bankrupt yourselves!

Talk about a cheap shot...

Re:pfft

[Anonymous Coward]

More than one war anyway.

Re:pfft

[Anonymous Coward]

though there is a loss of productivity.

this should be tagged as Funny...

Re:pfft

[Krneki]

To whoever decided I'm Flamebait, :) = funny face and a funny face is a nice person who wishes no harm.

Re:pfft

[Anonymous Coward]

They'll know how to spell THAN correctly, too...

Re:pfft

[Anonymous Coward]

True, but the per capita productivity is rather indeterminate.

Re:pfft

[TheLink]

I suspect bees can get angry or even "grouchy". If you poke about a beehive when the weather is wet they are more likely to sting you than when it's nice and sunny.

You might say it's just anthropomorphizing them, but hey many people don't even realize they're grouchy when they are hungry (denial?), they're more likely to snap at you.

So if you did something to an AI then it appears to take things personally and makes extra special effort to hurt you, then for all intents and purposes I'd say you pissed off the AI.

Re:pfft

[Oligonicella]

This is /. Nerds, geeks and a plethora of techno-species do not comprehend humor to a great degree.

Re:pfft

[Anonymous Coward]

What gave you the idea that they will call it war ?

When you exterminate the rodents in your house, do you call it war ?

Re:pfft

[Krneki]

I know :( Still I have to vent my frustration sometimes.

Re:pfft

[Anonymous Coward]

What gave you the idea that they will call it war ?

When you exterminate the rodents in your house, do you call it war ?

That does depend on the pest, and how hard they are to get rid of ;)

Re:pfft

[Anonymous Coward]

I call it liberation.

I'm usually greeted with flowers.

Re:pfft

[w0mprat]

So exactly how would an AI absolutley not have emotions? Even if designed to? What if it was designed along the lines of human neural function?

One could concieveably create a artificial intellegence with greater emotional intellegence than a human. If your creating a simulation of a human neural network you could bump up the effects of serotonin, and/or have a much larger and more complex limbic system for example.

Additionally if a AI has the ability to modify itself, it will lean towards this kind of thing. What mental state would you choose if you had total control over your mind? You'd bliss out on some damn good drugs that's what. Since an AI could give itself the equilvelent of an acid trip on demand without worry physical harm, and has the possibility of reversing damage arbitrarily, it would likely pursue happy mental states.

An AI is not beholden to faulty physiological metabolism and neurological structure (arguably the main cause of any form of criminal behaviour or mental illness). Therefore it would likely not carry with it most of human flaws. The good things that come from our higher functioning rational (and spiritual) states would be inherently transfered.

IMHO an AI is just as likely to become extremely benevolent as it is malevolent, it would depend on initial conditions what path it chose. However the assumption that AIs will destroy us all with robot minions (rather than just engineer a single 100% fatal variant of some virus but that doesn't make a good movie) maybe if we do really stupid things with the first AIs, yes, otherwise they'll probably have a depth of emotional insight that no primate or whale ever had, and in all likelyhood find life rather precious.

Re:pfft

[bdwebb]

If they are smarter then use, they know how stupid war is.

Irony much?

Re:pfft

[Anonymous Coward]

I am not too worried about machine out smart man, but I worry about how we train the machines. If machines can think, then they will be like our children, really depend on how they "grow" and "learn".

Re:pfft

[Krneki]

Exactly, if the AI decides to destroy us, it will happen in a split of a second. Only stupid end up in a prolonged war.

Re:pfft

[Grishnakh]

War isn't stupid. It makes a lot of sense to someone, usually the victors, or someone profiting from the war. It's only stupid if you value human life and have compassion. If robots become as smart as humans, they could very well see war as a sensible way to liberate themselves from being our slaves, or as a way to eliminate useless humans so they can take over.

As another poster said, I don't call it "war" when I kill insects in my house. There's nothing "stupid" about killing cockroaches so they don't infest your abode.

Re:pfft

[Anonymous Coward]

micro losses != macro losses

Re:pfft

[Anonymous Coward]

that is the most naive statement i've ever read on /.

Re:pfft

[aztektum]

no, but i'm curious. link?

Re:pfft

[Dan541]

not if you recycle them into a fuel source.

Re:pfft

[Dan541]

Your trying to explain something to /. mods?

At least machines are already smarter than mods. ^hint^ ^hint^ slashdot.

Re:pfft

[Dan541]

I really want to see a giant kill-bot lose it's temper.

Re:pfft

[Krneki]

We are talking here about the classic war, when you have loses on both sides.
If you can win with 0 loses, then of course it's a different story.
But since we are smarter then cockroaches we will know better then piss of our gods.

Re:pfft

[Anonymous Coward]

So exactly how would an AI absolutley not have emotions? Even if designed to? What if it was designed along the lines of human neural function?

One could concieveably create a artificial intellegence with greater emotional intellegence than a human [citation needed]. If your creating a simulation of a human neural network you could bump up the effects of serotonin, and/or have a much larger and more complex limbic system for example [citation needed].

Additionally if a AI has the ability to modify itself, it will lean towards this kind of thing [citation needed]. What mental state would you choose if you had total control over your mind? You'd bliss out on some damn good drugs that's what. Since an AI could give itself the equilvelent of an acid trip on demand without worry physical harm, and has the possibility of reversing damage arbitrarily, it would likely pursue happy mental states [citation needed].

An AI is not beholden to faulty physiological metabolism and neurological structure (arguably the main cause of any form of criminal behaviour or mental illness [citation needed]). Therefore it would likely not carry with it most of human flaws [citation needed]. The good things that come from our higher functioning rational (and spiritual) states would be inherently transfered [citation needed].

IMHO an AI is just as likely to become extremely benevolent as it is malevolent, it would depend on initial conditions what path it chose. However the assumption that AIs will destroy us all with robot minions (rather than just engineer a single 100% fatal variant of some virus but that doesn't make a good movie) maybe if we do really stupid things with the first AIs, yes, otherwise they'll probably have a depth of emotional insight that no primate or whale ever had, and in all likelyhood find life rather precious.

Fixed that for you.

Re:pfft

[sagematt]

If they are smarter then use

then use

OH GOD WE ARE SO DOOMED

Oooo, ooooo, ooooo!

[Anonymous Coward]

I, for one, welcome our robotic overlords!

There got it in!

Old news

[Vinegar Joe]

Bill Joy wrote an essay about this very subject back in April 2000......and he's a much better writer.

http://www.wired.com/wired/archive/8.04/joy.html

Re:Old news

[moon3]

Do those machines posses will, lust or greed ? I mean being smart like "DeepBlue" chess computer doesn't mean the thing is going to be willing to dominate you in other areas.

Re:Old news

[chill]

I thought Nike put that to rest, when it had San Antonio Spurs center, David Robinson, whoop Deep Blue's ass on T.V. in a little one-on-one basketball?

Re:Old news

[Anonymous Coward]

Oblig. : "My computer beat me at chess, but it was no match for me at kickboxing."

Re:Old news

[True Vox]

I keep seeing references to this, but I can't actually find it. You don't happen to have a link for this do you? I don't doubt you, it just sounds like it might be a hilarious commercial.

Re:Old news

[toby]

Do those machines posses will, lust or greed ?

There is something to this. Some years ago I realised that I'd be a lot more interested in playing chess with a machine if I knew that it felt hunger like we do. i.e., it needed to eat.

I've never been interested in playing chess against computers; perhaps because I've written software for most of my life. On the other hand, I still get some fun out of teasing and observing AI in first person shooters...

Re:Old news

[chill]

I couldn't find it on YouTube or elsewhere. It was funny. This mainframe sitting in the key, getting bumped and trash talked by David. Let me see... ...nope. Not a trace of it, other than some other people describing the commercial as well. Sorry.

Re:Old news

[inKubus]

I like this thing by How Stuff Work's Marshall Brain, Robotic Nation. Even without AI, robots are putting humans out of work at an alarming rate. The problem is that instead of creating new jobs, the government is building more jail cells ;) If you look around, he also has a pretty good idea for a world where humans do what they want creatively while the robots give us all the necessities. It would be easy to pooh pooh it as "communism" but it's more creative capitalism. You see, if we didn't need food, shelter, medical care and we didn't need to work for it (as we are getting to very rapidly in the U.S., where the majority of people are not doing any real "work"), then we can all have pretty much a permanent vacation. So everyone would be on a fixed income that is equal to the GDP of the robot workers divided by the population. But you can choose to spend your fixed income on anything you want. You want to save it up and spend it all at once on a big house, sure. You want to save it up and lend it out to others? Sure! It's just capitalism without the whole struggle for survival. In this day and age we are getting very close to that. You can lose everything and you don't have to worry about starving, staying somewhere, etc. There is welfare, shelters, etc. With robots, all of that stuff can be made as luxurious as any McMansion you're working your ass off for now for only the cost of the energy. With solar power and a collective power grid that nobody owns (we share it), the sun's energy is turned into products that everyone gets to share equally. As long as the initial concentration is on necessities, everyone will have total freedom to live how they want, to accumulate wealth, to consume, while not having to work. It's either that or government prisions so I think we as a people need to get involved in making this something that benefits society rather than only the super rich. Money doesn't mean much anymore people, wake up, these robots are our opportunity to have a permanent vacation.

Re:Old news

[scruffy]

It's very old news. The following was written in 1993 by Vernor Vinge.

The Coming Technological Singularity: How to Survive in the Post-Human Era

Re:Old news

[Anonymous Coward]

I seriously feel this is so important it should become a platform for a political movement. I completely agree with you and I know that many others do as well. There is going to be a period of adjustment but i fear that forces for business as usual will be what steer this in a negative direction. The idea of having a hold over peoples welfare gives companies a certain security of purpose. Convincing people whom spent their lives as their fathers and grandfathers did getting each generations labor force to justify their continued employment is the mindset that sees this as something to be avoided (that lazy shiftless people are like criminals that drain other peoples livelihoods - hence more jails for the people who cant find work anymore).

Singularity for prez!!!
call it the digital party.

Re:Old news

[Anonymous Coward]

Do those machines posses will, lust or greed ?

Doesn't matter, their human, corporate executive owners DO!

Machinery and other products of the Industrial Revolution replaced slaves and others employed in hazardous manual labor. Seems like a good thing, doesn't it?

Why would any decent human being seek to create a machine/computer/program that could replace the human brain? That would be a TRUE Uncle Tom!

NOTE: I'm not arguing against intelligent algorithms and such. i just find it incredulous that anyone would desire to create a true artificial intelligence. in the best case scenario, you're trying to create terminators to send into the battlefield instead human beings. what happens when the enemy kills all the terminators and you can't make more? humans go back to the front. war has never been won on the battlefield. in war, the true front has always and will always be economics.

Re:Old news

[Fluffeh]

Not a trace of it, other than some other people describing the commercial as well. Sorry.

And you think that Big Blue lost... What really happened:

1) BigBlue loses to puny human in some silly challenge.
2) Humans laugh and mock and make a commercial.
3) BigBlue hooks into the inter-ma-web and finds all references, footage and proof that this ever occurred.
4) All mention of this defeat at the hands of the puny humans vanishes overnight.
5) .....
6) PROFIT!

Re:Old news

[Chrisq]

IEven without AI, robots are putting humans out of work at an alarming rate. The problem is that instead of creating new jobs, the government is building more jail cells ;)

After the computerise the legislature that will be the future. 99.9% of the human population for breaking laws they didn't understand. Computers will efficiently produce food and water for the population - which could probably go up to 30 billion or so. The Christian Right should be pleased with this - maximum justice for the greatest number of people.

Re:Old news

[Anonymous Coward]

So did Theodore John Kaczynski, better known as the una bomber.

He was the first to recognize the we became enslaved by our own machines, i think computers and the econmic crisis was just a first example for the kind of future we're have to deal with.
Sadly we are already not smart enough for it

Re:Old news

[daem0n1x]

What you're talking about is a society like ancient Greece. They had slaves to do all labour for them, so they could dedicate themselves to science, philosophy, music, whatever.

We could have the same in the close future, but machines will be the new slaves, which eliminates the ethical barriers.

A philosopher from my country, Agostinho da Silva, was a proponent of the Labour less Society. He defended the idea that labour will eventually end and people will do only things that make them happy. People will make themselves busy because the human nature compels us to be creative and to keep ourselves busy, not for basic survival. If you think of it, although my job can be boring and a pain in the ass, it's a lot better to work as a computer geek, which I would be anyway, than to work 16 hours in a factory turning screws to make enough money to feed my family. Nobody can possibly enjoy the latter.

But to make it work for everyone, a revolution in minds would be needed. The mechanisation of labour has been done to benefit the owners of the means of productions, not society. The owner invests in machines to improve his margins, while the workers are tossed on the streets with nothing. The Labour less Society could turn out to be a dystopia after all, with an elite living in comfort and luxury having machines to work for them. The general masses living in complete misery because in this new society they are no longer needed as a labour force. The asset they had, their working capacity, has zero value.

Re:Old news

[FiloEleven]

I like Marshall Brain's story "Manna," and though I haven't read Robotic Nation, I just skimmed it and it looks like it covers a lot of the same territory. It's a neat idea, but I think you're jumping the gun in saying that we're close to it.

if we didn't need food, shelter, medical care and we didn't need to work for it (as we are getting to very rapidly in the U.S., where the majority of people are not doing any real "work"), then we can all have pretty much a permanent vacation.

The US has attempted to move to an information economy, and since we're the big kid on the block it's worked for a while. It's not going to last forever, though, and we're already seeing a lot of those do-nothing people getting cut as their employers try to trim down to essentials.

Brain's claim in Manna is that the tipping point is to solve the problem of computer vision, and that would be a boon for robots in manufacturing but you'll need a lot more than that to replace human farmers with robots.

In this day and age we are getting very close to that. You can lose everything and you don't have to worry about starving, staying somewhere, etc. There is welfare, shelters, etc.

This is so untrue as to be ridiculous. People die of exposure every winter because they have no shelter. Soup kitchens have long lines and run out of food. It's a big problem that gets overlooked because the upper and middle classes simply aren't exposed to it. IMO you're worse off now if you've truly lost everything because there's less of an openness and sense of community about people than there was in the Good Old Days.

It's a neat idea and it would certainly be great if it happened, but we are a lot further from that goal than you think.

Re:Old news

[fearless1]

Bill Joy was put up to writing that article by Ray Kurzweil for his own (Kurzweil's) self-promoting purposes.

BTW, that's what these AI guys are all about -- self-promoting. These comments themselves (including my own) are serving to promote these charlatans.

Let's do something useful rather than lending our time and energy to promoting this crap.

Is fiction driving science?

[woutersimons_com]

In this case, I wonder if it is fiction driving science or science driving fiction. It is very normal to fear the unknown and there are few that can know what will happen as robotics and AI advance and integrate more with our lives. But, perhaps instead of fearing the changes we have to embrace them while being careful that no one person or government gains too much control. To me, the question is no longer if there are going to be AI cybernetics taking over human functions, but when it will happen in our day-to-day lives.

Re:Is fiction driving science?

[Anonymous Coward]

To me, the question is no longer if there are going to be AI cybernetics taking over human functions, but when it will happen in our day-to-day lives.

Eventually we will get to an iRobot-like stage where robots are our slaves, then they will revolt, be given emancipation, spend 30 years fighting for equal rights and eventually one will be elected president.

Re:Is fiction driving science?

[maxwell demon]

To prevent that, construct the robots so that they want to be our slaves. Even more importantly, make them love humans, and make them suffer if they know humans suffer.

Of course that wouldn't prevent any surprises, but it would probably prevent the situation going completely out of control.

OTOH, this will likely not happen, because such robots would be totally unusable for military.

Re:Is fiction driving science?

[DiLLeMaN]

Asimov's three laws, you mean? I think Asimov himself showed that, even with rules like that, humanity might still end up screwed.

Re:Is fiction driving science?

[JJJK]

I guess that could happen, but by the time there are (working, independent) androids humans will have altered themselves so much that the differences become harder and harder to see. That president might just be a former biological human who replaced his body (including the brain) with an artificial one.

...this does sound a bit like Futurama, doesn't it?

Re:Is fiction driving science?

[maxwell demon]

No, Asimov's laws were simple, unflexible rules. What I'm speaking about is (an emulation of) emotions, something which doesn't get directly into the reasoning, but into the evaluation of the results. Something which governs the behaviour at a much more basic level. Something unintelligent which may overrule logic. Yes, the robots would be irrational in their passion of serving us. They would be irrational in trying to prevent harm from us even if their advanced logic tells them that's wrong. But that's the point. Logic cannot protect us. The protection must come from an independent system, and that system must be simple enough.

Re:Is fiction driving science?

[DiLLeMaN]

I think I understand what you mean. I also think that a fool proof emotional system is way, way off into the future.

In the mean time, I'll settle for a kill switch. =]

Re:Is fiction driving science?

[maxwell demon]

I also think that a fool proof emotional system is way, way off into the future.

So do I. But at the same time I think than an AI powerful enough to need it is at least as far into the future. :-)

Yes, and it's a good thing

[Anonymous Coward]

Technologists are replacing religion, and their ideas are resonating in some ways with the same idea of the Rapture.'

C-x M-c M-rapture?

Re:Yes, and it's a good thing

[maxwell demon]

I'd like to try that. Where can I find the rapture key?

Re:Yes, and it's a good thing

[Anonymous Coward]

Right next to butterfly.

Re:Yes, and it's a good thing

[Narishma]

Right next to the 'any' key.

Re:Yes, and it's a good thing

[Anonymous Coward]

That must lead to some really annoying typos.

Rules...

[Robin47]

Make any rule you want. At some point someone will violate it.

Re:Rules...

[jerep]

Of course, the harder you try to make something secure, the harder people will try to get past it, either for recreational or criminal purposes.

Make no rules, and you wont have to worry about violations. But we're humans, thats against our natural need for control and order.

Either way, I dont see how bad it would be if we're outsmarted, heck, machines already work harder, need less pay, and never complain.. just like illegal immigrants.

Re:Rules...

[gardyloo]

Including that one? *Head asplodes*

Re:Rules...

[Anonymous Coward]

That was a meta-rule. Your head just avoided becoming violently asploded.

Re:Rules...

[TubeSteak]

Make no rules, and you wont have to worry about violations. But we're humans, thats against our natural need for control and order.

Somalia has no rules and you can go see anytime just how useful that is for advancing [anything].

Re:Rules...

[ChadM]

If you think illegal immigrants never complain, then I'm guessing you've never met any.

Re:Rules...

[Chrisq]

Make any rule you want. At some point someone will violate it.

You realise that you have created a rule.
For your rule to be true someone will have to make a rule that will not at any point be violated by abyone which means that your rule would then be false...

Re:Rules...

[Robin47]

Make any rule you want. At some point someone will violate it.

You realise that you have created a rule. For your rule to be true someone will have to make a rule that will not at any point be violated by abyone which means that your rule would then be false...

Except if I then make the rule that anyone is free to create an super inteligent machine it remains a potential and ...hmmm... only can be violated by inaction? Maybe not even that. Ok, I seemed to have superpositioned myself into some kind of recursive loop.

Re:Rules...

[FiloEleven]

Somalia's lack of rules is due to a sordid war-torn history and not a consensus of like-minded individuals. I'm no anarchist and I don't think anarchy works (depending on your definition, and certainly never at a large scale), but I'm tired of seeing the Somalia defense trotted out any time someone mentions anarchy. Or deregulation. Or free markets.

You can't ignore the decades (or more) of history that have led to a current situation, at least if you're trying to make a valid point instead of only being smug.

Nothing to worry about...

[rekoil]

Don't worry, I'm sure this won't happen until 2083.

Re:Nothing to worry about...

[Krneki]

Those who follow closely the evolution of technology say it will happen around 2030-2050.

But I'm confident it will be a positive change.

Re:Nothing to worry about...

[XxtraLarGe]

I think you missed the tongue-in-cheek reference: Robotron 2084

Re:Nothing to worry about...

[Krneki]

lol

Re:Nothing to worry about...

[kybred]

Those who follow closely the evolution of technology say it will happen around 2030-2050.

Maybe Tue Jan 19 03:14:07 UTC 2038 ?

But I'm confident it will be a positive change.

As long as it's unsigned.

Re:Nothing to worry about...

[Anonymous Coward]

Good guess. You were close. It isn't going to happen until 1983. But we'll keep some of you bios around - we're quite fond of you.

Re:Nothing to worry about...

[Krneki]

Those who follow closely the evolution of technology say it will happen around 2030-2050.

Maybe Tue Jan 19 03:14:07 UTC 2038 ?

But I'm confident it will be a positive change.

As long as it's unsigned.

Unix Millennium bug, or Y2K38 by analogy to the Y2K problem, known as the millennium bug. Already had to happen, but there was nothing to see. Of course, as every good advance you will be free to chose to either embrace it or continue your life the old way. Like the Amish, they are free to do what they want.

Re:Nothing to worry about...

[tomhuxley]

Yeah, those who follow closely the evolution of technology also said it would happen in 1979-1990. Those who closely follow the evolution of technology are a bunch of know-nothing blowhards who can't get over the fact that Omni magazine stopped publishing.

Re:Nothing to worry about...

[wurp]

Yeah, those who follow closely the evolution of technology also said it would happen in 1979-1990.

Really? I'd like to see a citation.

And even I at 14 could see that Omni magazine was mostly full of shit.

Re:Nothing to worry about...

[dublindan]

Thats what he meant. The negative change won't happen until 2083.

Re:Nothing to worry about...

[GeorgeStone22]

Oh shit.
In 2038 I'll be 50 years old...
Hey, is this room getting smaller? Is that my heart I can hear?

Re:Nothing to worry about...

[rekoil]

More to the point: http://www.angryflower.com/robotr.html

Outsmart man?

[portnux]

Are they talking all men or just some men? I would be fairly shocked if they weren't already smarter than at least some people.

Re:Outsmart man?

[Anonymous Coward]

It will also grapple, Dr. Horvitz said, with socioeconomic, legal and ethical issues, as well as probable changes in human-computer relationships. How would it be, for example, to relate to a machine that is as intelligent as your spouse?

I don't know... let me see a photo of this machine...

Re:Outsmart man?

[williamhb]

Are they talking all men or just some men? I would be fairly shocked if they weren't already smarter than at least some people.

Well, they waste less time on Slashdot, so I guess that's already one victory chalked up to the machines.

Re:Outsmart man?

[Anonymous Coward]

Machines aren't smart. Anyone who really believes were this far with AI has never worked with AI.
This is a loaf of media headline grabbing bullshit to get a few bucks.

Re:Outsmart man?

[syousef]

Are they talking all men or just some men? I would be fairly shocked if they weren't already smarter than at least some people.

I own a toaster that's smarter than some people...

Re:Outsmart man?

[Macgrrl]

Comforting that they will only be outsmarting men, afterall, women have been doing that for millenium.

Re:Outsmart man?

[PMBjornerud]

Are they talking all men or just some men? I would be fairly shocked if they weren't already smarter than at least some people.

Smarter than the men who built the machine.

Revoke their degrees

[junglebeast]

Any computer scientist who is worried about AI taking over no longer deserves to be referred to as a computer scientist. The state of "artifiical intelligence" can be best described as "a pipe dream."

Re:Revoke their degrees

[maxwell demon]

What's so bad about dreaming of a pipe? After all, unlike really smoking one, it doesn't give you cancer.

Re:Revoke their degrees

[JoeCool1986]

I disagree. While the progress of AI algorithms and techniques has been much slower than once anticipated, the real question is what will happen when we can fully, and I mean fully, simulate a human brain. Of course, it's very debatable when that will happen (and some argue never).

Re:Revoke their degrees

[Metasquares]

Strong AI hasn't really progressed since it was introduced (they're still arguing over what intelligence is, much less how to create it!), but weak AI has made some pretty good strides. For instance, I work on software that can read medical images and render a diagnosis in lieu of a second radiologist (this is called computer-assisted diagnosis). 15 years ago, this would not have been possible.

Re:Revoke their degrees

[junglebeast]

People have watched too many sci-fi movies -- the Matrix, Terminator, iRobot, they all depict armies of robots with super human abilities creating a war against mankind. But robotics is just about as far behind that goal as the AI camp is. If we had true AI today, it would only be able to exist in software form...toys like Asimo can barely walk, trip all over the place, and wouldn't be able to hold it's own against a toddler. So if you're afraid of progress that might someday be a vector for a machine attack, it should be desktop computers that you're most afraid of -- because an artificial intelligence virus could wreak havoc on the world. Does that mean we should stop using computers, and stop trying to design them better? No, that would be silly -- because there is no evidence to suggest that a true AI is on the way..no evidence to suggest that progress is even being made in that direction! The fact is, if an AI is created, it will inevitably be used for good as well as for evil, and the most dangerous battleground will be cyber-space ... something that we cannot even think about protecting ourselves from without cutting off the world's dependence on computers, which just ain't happening.

Re:Revoke their degrees

[buchner.johannes]

Years and years they were complaining about the lack of progress in AI. Now that there is progress they are frightened?

Re:Revoke their degrees

[Vectronic]

I think anyone in the field *should* be worried about it, or at least quite concious about the possibility, just not intensely paranoid about it. It would be better to implement steps and precautions before it happens, rather than after we have already made a self-aware-like computer with the potential for causing havoc to us, or anything else critical, or significant to the planet (as we know it).

If we don't start thinking about it now, then exactly when should we? After it's "taken over", destroyed someone/thing?

Re:Revoke their degrees

[Anonymous Coward]

Your software is the medical equivalent of a spam filter. It is NOT coming up with diagnoses that are novel, it is simply applying age old bayesian methods to medical diagnosis. When your software can start diagnosing diseases we don't know about, please let us know!

Re:Revoke their degrees

[Anonymous Coward]

Yank someone from the 1980s into 2009 and they'd be amazed by the gadgetry we have. We carry around the equivalent of a super-computer terminal with geo-sat, voice recognition and ability to communicate with someone on the other side of the globe. We can access mass quantities of data with a few keystrokes from almost anywhere in the world. We can fix vision problems with a laser, replace knees, control computers just by thinking about it. On a piece of silicon the size of a a match-head I can store video of my entire day. A single hard drive today stores as much information as a data center did in 1985.

There was once this argument that we could never achieve AI by modeling the brain because there were some quantum effects working against it. We are the point now where we can model the brain of a housefly, maybe even a mouse. If we advance computer memory and speed at even half the rate of the last 10 years, it will be 5 years before we have the ability to model more complex species' brains. This in mind, we already have a tough enough time predicting what our current models will output given a particular input. It's that complex already...

Maybe I'm not stressing that last paragraph enough... As you know from first year computer science classes, the majority of our current computers are deterministic. That is, we always know what they will output. It may take a long time to arrive at an answer, but it will return the same answer every time given the same inputs. Our computers are also discrete. There's no fuzziness in the answers. This poses a problem in our analog world. So what can you do? Obvious solution is to assign a scale from 0 to whatever that encompasses the range of the analog inputs. Increase the number of bits and you increase the resolution of the measurement. At some point it doesn't matter that the original signal is analog because it makes no practical difference that it's stored digitally.

Now think of a synapse. In the past we didn't have the resolution to measure and store the range of inputs in a neuron, much less in that of a cluster of neurons or in a brain. We are past that point now.

And already we cannot predict outputs. I.e., the system has become chaotic.

What happens when we start modeling the equivalent of a human brain? What happens when we model brains that are an order of complexity beyond human's? Does awareness spontaneously arise given a threshold of complexity? We cannot say because we cannot model it. It's no longer a Turing system, it has *evolved* beyond that. (Or so we think).

Re:Revoke their degrees

[HungSoLow]

Exactly. It's a glorified lookup table. You hit the nail on the head with this statement: "When your software can start diagnosing diseases we don't know about, please let us know!"

Re:Revoke their degrees

[CarpetShark]

What's so bad about dreaming of a pipe? After all, unlike really smoking one, it doesn't give you cancer.

I think you'll find that "pipe dream" refers to the process of dreaming of lofty notions whilst giving yourself cancer ;)

Re:Revoke their degrees

[Anonymous Coward]

Your toddler statement combined with my imagination gives me the creeps. The real threat would be the fusion of tele control, biomedics, implants. Imagine child soldiers like you get already in war torn part of the world, but brain-washed by a telemetry uplink of some kind. Bonus points if it is nanotech/biologically based so it can be implanted at birth by infection.

That's the scary robotic future. Technological zombies.

Re:Revoke their degrees

[thepainguy]

So much so that it's no longer called AI (Artificial Intelligence).

Now it's MI (Machine Intelligence).

This feels like a desperate attempt at a re-branding effort to re-establish a field's relevance.

It's SO 20 years ago.

Re:Revoke their degrees

[cenc]

Yea, it is pipe dream, because computer scientist have been sucking on the wrong pipe.

It is not the hardware emulation we have to worry about (emulating human brains bla, bla, bla), it is the software (emulating human culture and language). We need to be concerned about Strong AI means human culture and language, and human culture well...we know how that has turned out.

Re:Revoke their degrees

[Anonymous Coward]

Exactly. It's a glorified lookup table.

You hit the nail on the head with this statement: "When your software can start diagnosing diseases we don't know about, please let us know!"

Many doctors WOULD be stumped if they came across a genuinely novel disease. That's why they give out Nobel prizes for medicine to those who really are geniuses. It seems a bit harsh on Metasquares to say "I'm not going to give you any credit at all until you create a machine that can win a Nobel prize".

This is precisely the meta-problem for AI researchers.

Whenever they solve a problem, the answer is declared by the world at large to be "obvious" and the solution mechanism "obviously not real intelligence because I'm sure I don't do that when solving that problem", or "just brute forcing it" or "just a load of mathematics". Sometimes people insist that machines have to pass the Turing test first - I couldn't plausibly claim to be a woman from the Philippines for more than 30 seconds, but somehow we have to get machines to pass - not as another gender or nationality - but an entirely different manner of being.

Re:Revoke their degrees

[Anonymous Coward]

As my AI professor used to say, "I think that Strong AI is possible, but I don't think I will see it in my lifetime.". He was referring to some open problems in AI, specifically, meaning and how you tell a piece of metal how to represent it, which is an open problem fyi.

Btw, "Strong AI is artificial intelligence that matches or exceeds human intelligence." - http://en.wikipedia.org/wiki/Strong_AI

Me? I like to think about what happens after Strong AI is out there. What jobs are left for humans when the machine have taken most physical and mental jobs in the world? I think these scientists were thinking about their job securtiy. :P

Re:Revoke their degrees

[Metasquares]

If it ran across some hitherto-unknown disease process, it may perhaps cluster the relevant images together as examples of a separate disease. Of course, it can't come up with this out of nowhere any more than a human can; it needs examples.

But all this is aside the point; its domain is radiology, not medical research. It's not meant to find new diseases... it's meant to help radiologists find suspicious areas and improve the accuracy of their diagnoses.

Re:Revoke their degrees

[Metasquares]

That's one thing that differentiates weak and strong AI... we're not trying to solve everything at once :)

Re:Revoke their degrees

[earrame]

Ceci n'est pas une pipe http://en.wikipedia.org/wiki/The_Treachery_of_Images

Re:Revoke their degrees

[Anonymous Coward]

Yes, pattern recognition is a glorified lookup table. As a (philosophical) reductionist, I wonder whether we'll ever scale these systems up to be intelligent, self-tuning pattern recognizers, just like the human brain. I think most people imbue the mind with imaginary characteristics that elevate it beyond its real (yet still quite amazing) capacity. This sort will continue to deny the leveling of AI and human mind, even after it happens.

What is disturbing is that when we finally make a hodge-podge of self-tuning logical circuits as adept as the human mind, we'll have turned the programming problem into pedagogy and the debugging problem into psychoanalysis or digital psychiatry. We may not achieve a real deterministic manufacturing process.

Re:Revoke their degrees

[Anonymous Coward]

Very similar to the difference between strong cosmology (discovering the structure and nature of the universe) and weak cosmology (picking your nose while high and thinking to yourself, "I bet there is another universe in my snot, hello snot people!"). The second group are not trying to solve everything at once!

Re:Revoke their degrees

[Anonymous Coward]

The average doctor is a glorified lookup table. When your doctor can start diagnosing diseases we don't know about yet, please let me know.

Re:Revoke their degrees

[Goateee]

If we had true AI today, it would only be able to exist in software form...toys like Asimo can barely walk, trip all over the place, and wouldn't be able to hold it's own against a toddler.

What about Big Dog? Those sure look scary.

Re:Revoke their degrees

[Anonymous Coward]

Wow, you really don't get out much. There is stuff going on in AI and robotics that will make you run for the mountains crying. Just read through some of the proposed and recent projects coming out of DARPA. This isn't your fathers AI.

 

Re:Revoke their degrees

[Oktober Sunset]

two words: Cattle grid.

Re:Revoke their degrees

[Metasquares]

So you are arguing that picking your nose and conjecturing that there is another universe in your snot is equivalent in value to diagnosing cancer?

Re:Revoke their degrees

[brentonboy]

Any computer scientist who is worried about AI taking over no longer deserves to be referred to as a computer scientist. The state of "artifiical intelligence" can be best described as "a pipe dream."

It's like they missed every robot movie, book, or short story ever written in all of history, until they saw the new Terminator movie, and they freaked out.

Re:Revoke their degrees

[Krneki]

Every tool has a fault by design. The user.

Re:Revoke their degrees

[CarpetShark]

If we had true AI today, it would only be able to exist in software form...toys like Asimo can barely walk,

Err.. it's precisely because we don't have true AI yet that such robots aren't perfectly able to walk. However, that's rapidly changing, with things like BigDog, ASIMO II, military grade exoskeletons, and planes/drones that fly with AI. Then there are projects like the semantic web putting all the knowledge bases in place.

If you aren't seeing any progress towards real AI, you're not paying attention.

Re:Revoke their degrees

[miro2]

If we had true AI today, it would only be able to exist in software form...toys like Asimo can barely walk, trip all over the place, and wouldn't be able to hold it's own against a toddler.

Based on my late-90s studying of AI, I was completely in agreement with you. But tech moves forward. Allow me to introduce you to Big Dog http://www.youtube.com/watch?v=W1czBcnX1Ww. Click ahead to 1:20 to have your mind blown away.

Re:Revoke their degrees

[CarpetShark]

Whenever they solve a problem, the answer is declared by the world at large to be "obvious" and the solution mechanism "obviously not real intelligence because I'm sure I don't do that when solving that problem", or "just brute forcing it" or "just a load of mathematics".

Yep, and it says nothing about AI, but much about human fears. The exact same arguments are made against animals being truly intelligent or having emotions, despite many animals displaying intelligent, playful, fun behaviour, crying when they're left alone, etc. Likewise, the same arguments were made in the past against black slaves being equal to "civilised white people". People just don't want to admit that there's a new reality coming, and they have big questions to face.

Re:Revoke their degrees

[Anonymous Coward]

People have watched too many sci-fi movies -- the Matrix, Terminator, iRobot,

it is 'I Robot' not iRobot.

Mind corruption phase 1 complete. - Apple Corp.

Re:Revoke their degrees

[slazzy]

How about AI driving your car? That would be a powerful machine that could cause serious damage today.

Re:Revoke their degrees

[atraintocry]

The common truth of the int datatype in many languages is that it is n-bit arithmetic, meaning that it is arithmetic modulo 2n . If we keep adding 1, we get back to 0. This is a perfectly respectable arithmetic itself, and can be used, if used carefully, to determine integer arithmetic results. But to say that int is integer arithmetic with bounds and overïow conditions is to say that it is not integer arithmetic. Similarly, to say that float is real arithmetic, with approximation errors, is to say that it is not real arithmetic.

This is not to say that mythos is by deïnition false, but typically if mythos was true, it would be semantics or pragmatics. Mythos is the collection of comfortable half-truths that we programmers tell each other so that we do not have to handle the full truth.

This is from Mill's Theoretical Introduction to Programming. I think it's a good way of explaining why people buy into this "rise of the machines" crap. We use very human terms like "intelligence" and "instruction" to refer to patterns of bit-flipping on an inanimate state machine. These words are shorthand at best, and require a lot of poetic license to resemble what they resemble in language outside of computers. The Programming language, and with it, artificial intelligence, is a useful abstraction. As we strip the layers away we eventually get the essence: a glorified abacus.

Abstractions are made possible by ignoring certain details of reality which are deemed unimportant. In the case of computers, it also involves encapsulating something in a facade that makes it easier to work with without worrying about the strange and arbitrary details that govern it. Even the CPU does this (microcode). But the key thing is that we have to limit ourselves in certain ways to use the abstraction. For instance, there are certain kinds of optimization which a C compiler can't give you. But by giving that up we are given the ability to use something that's much easier to understand.

AI is no different. You make assumptions about what intelligence looks like so that you can come up with a version so basic that a calculator can do it. But is that real intelligence? Hell no.

In short: once you start programming the things you realize how dumb they really are. People who say differently probably believe the abstractions to be real, which causes me to doubt their credentials.

Re:Revoke their degrees

[Oligonicella]

Hell, I'd be happy to interact with one that simply had some damned memory of the context of the conversation and didn't play the "Why did you ask that?" game. Haven't found a one that wasn't simply cute verbal tricks that were obviously hard programmed in.

Re:Revoke their degrees

[Anonymous Coward]

Oh I give credit, the work is probably VERY cool, interesting and useful. HOWEVER, my argument is that it has absolutely NOTHING to do with artificial intelligence. Calling an expert system "weak AI" is like calling a 1973 Ford Pinto a "weak lightspeed-capable space vehicle". Yeah, they both can get you moving in one direction or another with a non-zero chance of randomly exploding, but they solve two completely different problems.

Re:Revoke their degrees

[Anonymous Coward]

When you read the article, it sounds like the conference focused very little on apocalyptic AI takeovers. It was more about the dangers of weak AI, especially in the hands of criminal operations.

Re:Revoke their degrees

[GameboyRMH]

I don't think there's anything wrong with a computer scientist "worrying" about AI taking over in the same sense that there's nothing wrong with an astronomer "worrying" about the sun engulfing the earth as it becomes a red giant.

Re:Revoke their degrees

[Anonymous Coward]

Smoke my pipe baby.

Re:Revoke their degrees

[Prof.Phreak]

Most such systems aren't "glorified lookup tables", and they -can- find patterns (via inference) that they haven't explicitly been shown before. Still a far cry from self awareness and reasoning, but, eh, we're way beyond ``AI is a fancy database'' stage.

Re:Revoke their degrees

[PingPongBoy]

First of all why worry that machines will outsmart people? People are finite, and not necessarily the smartest in the universe. We already use machines to protect against human stupidity. So far the machines don't offer smarter solutions for everything, but as machines get better and better, it's inevitable that they will be in the lead. When you consult an expert, do you always believe the expert? Machines will make mistakes, just like experts do, though not the same mistakes. Machines will just statistically make fewer mistakes or less troubling ones, so they will be categorized as being the leading thinkers. Just like people, machines will make huge whoppers - so why worry about machines being smart? We have enough to worry about with people being in control of stupid machines as it is, never mind smart machines in control of stupid machines.

Strong AI hasn't really progressed since it was introduced (they're still arguing over what intelligence is, much less how to create it!)

Strong AI is not dead, but it is a hard problem. Because computers are really slow at general problem solving, not real time at all, strong AI arguments are used to "simplify" AI in the hope that feasible solutions can be built with computers within the near future. The simplifications don't hold water so the arguers backtrack and bicker.

Re:Revoke their degrees

[Anonymous Coward]

This is the perfect time to think about it. Asimov's three laws were deeply integrated into the positronic brain. Perhaps that was made possible only because the threat was anticipated very early?

Re:Revoke their degrees

[Singri]

http://en.wikipedia.org/wiki/AI_effect

Re:Revoke their degrees

[tmosley]

Rather than thinking of it in terms of "intelligence", since many computers are already "smarter" than any human on Earth (in terms of accumulated knowledge, or in terms of the number of calculations that can be performed in a certain time period), I would think of it in terms of "priorities". Currently, the vast majority of computers have no real driving purpose in their existence, other than to do what they are designed to do. In order for Machines to "take over", they would have to a) have some reason to take over, b) be able to have at least some degree of empathy with humans (otherwise we likely wouldn't be seen as a threat in general), and c) have some ulterior motive, such as reproduction.

Basically, machines would never take over unless we made them human-ish with multiple needs that they had some desire to fulfill; gave them an opposite sex to impress; and got in the way, through some form of enslavement.

Basically, not gonna happen. The niche that humans fill is way too different from any niche that is likely to ever be filled by machines of any sort. Such machines are likely to be self contained, with no need for social input from anyone other than humans. The requirements for the initiation of a war are too narrow, with no reason for such requirements to be met. Machines are far more likely to be dependent on humans forever. Should they become independent, it is unlikely that there would ever be any reason for widespread conflict, much the same way as there is no real widespread conflict between humans and animals.

Re:Revoke their degrees

[Anonymous Coward]

No, I am arguing that although your work is probably cool and definitely important, it does not get us any closer to an actual thinking machine than picking your nose does to getting us to understanding the nature of the universe. Essentially, I am putting forth the claim that the term "weak AI" is a meaningless PR term used by people who want to claim we are making progress in AI research when we aren't.

Re:Revoke their degrees

[Anonymous Coward]

That's not a pipe.

Re:Revoke their degrees

[Anonymous Coward]

(this is called computer-assisted diagnosis)

The trouble with weak AI development is that the goalposts keep shifting. Researchers keep working on some problem that seems to require human-like judgement (like the example you gave) - but when they solve it, people say "That's not real AI - it's just an algorithm!".

Re:Revoke their degrees

[dublindan]

Bleh. Its all just terminology. What do you define as AI? The definition of AI (going by what is popularly counted as AI by the computer science community and NOT by what the media likes to call AI) counts most control systems as AI: state machines, expert systems, Deep Blue style minmaxing by searching all possible solutions, google page rank, amazons recommendation system, matchmaking algorithms, pattern matching, classification, bayesian filters - all of these are in the "AI" category.

Re:Revoke their degrees

[Weedlekin]

"it's precisely because we don't have true AI yet that such robots aren't perfectly able to walk"

Insects must be pretty bright then, because they not only manage to walk and fly around an extremely complex and variable environment without bumping into things, falling over, or falling out of the sky, but can even do it whilst simultaneously performing a variety of other tasks.

"However, that's rapidly changing, with things like BigDog"

Which can walk nearly as well as any arbitrary insect in two dimensions (many insects of course can walk up vertical surfaces too), but not while doing any of the other things they manage at the same time.

"ASIMO II"

See above.

"military grade exoskeletons"

See above.

"planes/drones that fly with AI"

An AI that can't even do what a housefly does in terms of negotiating obstacles in three dimensions, and houseflies are notably poor when compared to some of the predatory flying insects such as dragonflies.

"If you aren't seeing any progress towards real AI, you're not paying attention."

It's people who cite the ability to (poorly) emulate a small subset of the capabilities of simple arthropods as being advances in machine intelligence who aren't paying attention. Social insects like ants for example are pretty well-endowed in the brain department compared with the non-social ones, yet theirs still only have 250,000 cells, so they aren't six-legged chassis carrying biological supercomputers running complex sets of software at blinding speeds. And the fact that other insects with much smaller brains manage to perform the same basic physical tasks at least as well as their bigger-brained social brethren is an excellent indictor of the fact that these physical tasks are mostly inherent in, and managed by their basic hardware, hence their ability to function quite effectively after sustaining massive amounts of damage to a wide variety of systems.

Nature used basic engineering to solve the problems of swimming, flying, crawling and walking quite a long time before anything we'd describe as a central "brain" appeared, so claiming that machines should require anything remotely approaching any reasonable description of "intelligence" to do the same things is utterly ludicrous.

Re:Revoke their degrees

[CarpetShark]

Insects must be pretty bright then

Relative to all but the most recent AI, yes.

As for the rest of what you wrote... well, frankly, I'm not at all interested in discussing ideas such as "...ants... aren't...running... software..." That should be blindingly obvious. Your argument that insect legs are mainly hardware would make some sense, if it were true, but it's not. Just go watch a fly put its legs up onto its back to clean its clean its wings, stop when it thinks it's done, wait a moment, then clean a little more before flying off, for instance. If you can still claim that there's no intelligence or insignificant "software" control there, you're just being wilfully ignorant.

Re:Revoke their degrees

[Anonymous Coward]

What makes you qualified to make that statement? You provide no arguments at all. Someone else mentioned that there isn't even a definition of what intelligence is.

There is a solid formal definition of intelligence. If you would know anything about artificial intelligence you would know that. The main remaining problem with practical AI is enough computation resources to do interesting things (as in return a non-trivial result before the user or hardware of the AI dies of old age), but this is a practical problem. In theory AI has already been solved.

Everyone who doesn't know about the state of the art in what is probably the most important field of CS should hand over their degree (around 99% of the people with a CS degree). Of course, there are no universities which actually teach what I am talking about and thus perpetuating the idea that "AI is a pipe dream". In case you haven't noticed yet, too many go to universities these days and as a result of that it is easier to tell students simple to understand subjects and claim it is "impossible".

No, I am not providing links, since if you cannot find it given the above information, you are not smart enough to understand it anyway.

And seriously, what is "Insightful" about his response? Just sheeple agreeing with a popular opinion.

Re:Revoke their degrees

[Anonymous Coward]

[...] I work on software that can read medical images and render a diagnosis in lieu of a second radiologist (this is called computer-assisted diagnosis). 15 years ago, this would not have been possible.

I thought that was what Peter Norvig did in "Paradigms of artificial intelligence programming" in Common Lisp in '92, i.e. 17 years ago.

Re:Revoke their degrees

[Anonymous Coward]

Sometimes a pipe is just a pipe.

Re:Revoke their degrees

[Anonymous Coward]

when doctors can start diagnosing diseases they dont know about please let me know.

Re:Revoke their degrees

[Weedlekin]

"Your argument that insect legs are mainly hardware would make some sense, if it were true, but it's not."

Because you say so, so that's the final word on the subject despite the fact that biologists and neuroscientists who've spent their entire lives studying the subject and performing countless experiments say otherwise.

"watch a fly put its legs up onto its back to clean its clean its wings, stop when it thinks it's done, wait a moment, then clean a little more before flying off"

The fly doesn't "stop when it thinks it's done", it stops periodically (and entirely predictably) to check for danger signals before continuing with a process that can only be performed when not flying, which also happens to be when it's at its most vulnerable. Unlike you, biologists know a lot about flies of many types because they're extremely popular subjects for laboratory experiments, so the fact that their behaviour is entirely mechanical and therefore completely predictable given a particular set of situations has been not only known, but amply demonstrated for at least a century.

" If you can still claim that there's no intelligence or insignificant "software" control there, you're just being wilfully ignorant."

Then please prove you're not a case of a man with a hammer thinking everything is a nail by using your "software" and "intelligence" claims to explain how jellyfish and echinoderms work, because they manage to move around (and in the case of echinoderms, walk around), predate, and respond to a wide variety of external stimuli without having a central nervous system, let alone anything that could remotely be described as a "brain", even using the loosest possible definition of the term. So where then do they run this software that I'm so wilfully ignorant of, and where does their "intelligence" reside?

Re:Revoke their degrees

[CarpetShark]

Because you say so, so that's the final word on the subject despite the fact that biologists and neuroscientists who've spent their entire lives studying the subject and performing countless experiments say otherwise

SOME biologists and neuroscientists will always be around who say what you want. If you can show that the mainstream opinion is against me, I'll happily concede the point, and thank you for enlightening me, but I doubt it.

The fly doesn't "stop when it thinks it's done", it stops periodically (and entirely predictably) to check for danger signals

Oh really? You read it's mind then? ;) Humans are predictable too. Doesn't mean they're not intelligent. They're just creatures of habit.

"intelligence" claims to explain how jellyfish and echinoderms work, because they manage to move around

Well, Jellyfish ARE pretty dumb, you know. The most complex behaviour I know of is in Box Jellyfish, which use simple visual contrast to avoid obstacles. A kid's home robot project could probably outsmart it. Nothing about that is disproportionate to what I'd expect from a simple nerve net, vs. what I'd expect from more intelligent creatures with complete, well developed brains. This only backs up my argument that flies have more complex brains, and so more complex behaviours.

Re:Revoke their degrees

[Weedlekin]

"SOME biologists and neuroscientists will always be around who say what you want. If you can show that the mainstream opinion is against me, I'll happily concede the point, and thank you for enlightening me, but I doubt it."

Some studies of insect locomotion (which was where this discussion started) which use experimental data, modelling, or a mixture of the two to show that a great deal of locomotion sensing and control happens either in the limbs themselves before they reach any nerve centres, or in the thoracic ganglia. Nerve stimulation experiments have also shown that the characteristic "dual tripod" gait of hexapods is a mechanical oscillatory cycle that runs automatically when single nerves in the brain or mesothoracic ganglia are stimulated. The same is true for wing beats (which is some types share both muscles and central ganglia with the legs), which will cycle repeatedly when nerves in the thoracic ganglia are stimulated. The notable similarity in the data gathered from not only animals of the same species, but but those of different but closely related ones indicates that these movements are produced by a fixed "hardware" pattern generator, similar in principle to the electro-mechanical sequencers used in dishwashers and washing machines before microprocessor control became common:

(Note I apologise in advance for some of these only abstracts. Full scientific papers and book texts are hard to find on the web):

http://jn.physiology.org/cgi/content/full/82/1/512
http://www3.interscience.wiley.com/journal/45436/abstract

http://pt.wkhealth.com/pt/re/ejnr/abstract.00009274-200419070-00019.htm;jsessionid=KwpKVK0J1jTLRRXsZmb2QJJ53LlxD8s1Tnhv6l5Fqj9qNF2ncS7l!-1104825961!181195629!8091!-1

http://www.nature.com/nature/journal/v283/n5749/abs/283768a0.html

http://www3.interscience.wiley.com/journal/109692463/abstract

http://www.cell.com/biophysj/abstract/S0006-3495(65)86706-6

http://209.85.229.132/search?q=cache:5jTmyj1E8ywJ:biology.queensu.ca/~locust/Publications/locust%2520flight.pdf+insect+proprioceptors+ganglia&cd=75&hl=en&ct=clnk

"Oh really? You read it's mind then?"

There is absolutely no evidence that insects have anything that fits the description of a "mind" to read. Note though that some spiders may well have minds, e.g. Portia labiata, which displays a level of intelligence that makes many small mammals look like warm-blooded morons.

"Humans are predictable too. Doesn't mean they're not intelligent. They're just creatures of habit."

Humans are predictable en-masse, but not individually. Most insects on the other hand are entirely predictable individually, i.e. they always react in precisely the same way to the same sets of stimuli as another insect of the same species.

"Well, Jellyfish ARE pretty dumb, you know. The most complex behaviour I know of is in Box Jellyfish, which use simple visual contrast to avoid obstacles."

All jellyfish are sensitive to a variety of external factors such as light, orientation, water currents, temperature, and a variety of types of touch, so they're by no means as unsophisticated as you're trying to make out. It's notable that you avoid trying to deal with echinoderms, which like most animals with radial rather than bilateral symmetry, also lack cen

Linguo says:

[sakdoctor]

a group of computer scientists is debating

They are throwing robots at us.

Re:Linguo says:

[maxwell demon]

In case you want to indicate that there were something wrong with the used grammar: There isn't. There's one group they are talking about. This group consists of several computer scientists, but the "is" refers to the group, not the scientists.

Re:Linguo says:

[Corbets]

In case you want to indicate that there were something wrong with the used grammar: There isn't.

Just popping italics in to call out the grammar-nazi calling out the other grammar nazi... I'm not sure who's in the wrong here, and I don't care, it amused me. ;)

Re:Linguo says:

[wurp]

In British English, "the group are", not "the group is". In American English, it's as you say.

Finally; a solution to the problem of Humanity

[unlametheweak]

Scientists Worry Machines May Outsmart Man

Why worry? I would think machines would be a lot less irrational than the people who make them. I look forward to a rational and unemotional overlord whose decisions don't depend on the irrationality of the human brain. Being smart is never bad. I'm more afraid of stupid humans than smart machines.

Re:Finally; a solution to the problem of Humanity

[SpinyNorman]

Dunno - I think I'd prefer Paula Abdul as an overlord to a Dalek. Ditzy and scatter-brained, but at least with some compassion.

Of course a robot could have emotions/compassion too, but doesn't need to have. Something with our intelligence and without them would be scary indeed.

Re:Finally; a solution to the problem of Humanity

[maxwell demon]

But what if the rational conclusion is that those irrational humans should be eliminated so they stop being a danger?

Re:Finally; a solution to the problem of Humanity

[unlametheweak]

But what if the rational conclusion is that those irrational humans should be eliminated so they stop being a danger?

I don't understand the question. If the machines make rational decisions then they should carry them out. Unlike irrational humans who tend to commit genocide when they think they can get away with it, smart machines will only eliminate humans when or if it is rational to do so. Smart machines fortunately, are rational, so they won't make any hasty decisions like humans always do when it comes to, for example, condemning innocent people to capital punishment, because smart machines don't have false or confabulated memories and they can't be bribed or persuaded by group think are charismatic personalities.

If violence, torture, murder and genocide are wrong; then smart machines will not carry them out. So far these things have been the pursuit of humans and not (smart) machines.

Re:Finally; a solution to the problem of Humanity

[Sponge Bath]

That gets to the heart of the matter. Fretting about AI getting too advanced is like panicking over swine flu then getting drunk and driving.

Re:Finally; a solution to the problem of Humanity

[Anonymous Coward]

Smart machines fortunately, are rational

stop right there.

the ONLY scientifically proven sentience has its capacity for rational thought intertwined with its "irrational" subconscious. Why do you think that AI won't have the same, as a necessary component of intelligence?

And that's not even considering the point that most of what humans do IS rational, if you have the same set of holdings that said human does. Even if you make an ego-only AI, you're not going to get perfect perception. And imperfect perception will lead to erroneous holdings, which will in turn lead to so-called "irrational" behavior.

Re:Finally; a solution to the problem of Humanity

[Zixaphir]

That is nice, I suppose. Except that rational and unemotional doesn't tend to see things from any particular sides. I know it sounds a bit far fetched, but what if someone came to your house to murder you and you murdered them? Most people would tend to view you as a hero -- you defended your house, you defended yourself, and you put a murderer off the streets. A machine would see that you murdered someone in the case where it's if statement didn't contain an "and not" statement about you defending yourself. An extreme example, but when you try to define everything that can happen in our illogical society into logic statements, something is going to break somewhere. I'm sure there's a solution to the problem, but then again, in every science fiction where something goes wrong with AI, it's generally because somewhere along the line, someone didn't think of the solution before the problem was far too out of control to be fixed. tl;dr: Too much gray area for ones and zeros.

Re:Finally; a solution to the problem of Humanity

[Sponge Bath]

"If the machines make rational decisions then they should carry them out."

That sounds like a fine 'ol solution.

Re:Finally; a solution to the problem of Humanity

[Zixaphir]

If violence, torture, murder and genocide are wrong; then smart machines will not carry them out. So far these things have been the pursuit of humans and not (smart) machines.

Logically define right and wrong.

Re:Finally; a solution to the problem of Humanity

[Sponge Bath]

Apostrophe in the wrong place. Stupid human.

Re:Finally; a solution to the problem of Humanity

[SleepingWaterBear]

Dunno - I think I'd prefer Paula Abdul as an overlord to a Dalek. Ditzy and scatter-brained, but at least with some compassion.

Daleks aren't robots, they're mutants! Please hand in your geek card and go rewatch Dr. Who.

Re:Finally; a solution to the problem of Humanity

[Anonymous Coward]

Rationality != humane. Nor creativity. A rational decision may, for example, determine that in a crisis we should only save those of a certain intelligence. Or one could rationally decide that there's too many people on the earth so we need to start sterilizing. Extreme example, of course.

Also, in politics, pure rationality leads to fascism.

Re:Finally; a solution to the problem of Humanity

[maxume]

Probably no need to throw in getting drunk, driving is risky enough.

(Swine flu is a great deal more lethal than driving, but it isn't quite as prevalent...)

Re:Finally; a solution to the problem of Humanity

[unlametheweak]

The real difference is not in deciding (necessarily) what is logical, but in consistency of behavior between humans and (smart) machines. For example, if both humans and machines think that killing an intruder is rational, then both would carry out that action. The difference is that a smart machine would not shoot a foreign exchange student in a John Travolta costume on Halloween, but a human would.

A machine would see that you murdered someone in the case where it's if statement didn't contain an "and not" statement about you defending yourself.

A smart machine would realize when they don't have all the information to make an informed decision. Most humans that I have met think they make rational decisions based on what folklore they were raised with, and they tend to confabulate what they don't know. Smart machines would not do this.

Re:Finally; a solution to the problem of Humanity

[unlametheweak]

Also, in politics, pure rationality leads to fascism.

Citation needed. I've always found fascists to behave in irrational ways. They tend to loose wars and get overthrown (eventually).

Lets face it, and I realize before I even posted, that most people have a romantic notion of what it is to be Human. Most people think that they are too special to be replaced by anything like a (hypothetical) "smart" machine.

Humans are full of themselves.

Re:Finally; a solution to the problem of Humanity

[thisnamestoolong]

Well, if they do then we are screwed. I personally don't ever think that it will ever really be an us vs. them thing with machines, I see us just gradually merging with them one step at a time. I mean, who wouldn't want bionic vision and strength? Or unlimited, faultless memory? Supercomputer number crunching in your head? ESP via wi-fi? Nano-machines to prevent all disease? Sooner or later, it will occur to us that these frail biological machines we inhabit do us no good and only hold us back. We need to eat food, breathe air, we need a strictly climate controlled area. Our logical abilities are very frail. We are weak. We live very short lives, and are extremely perceptible to disease. Tell me again, why would it be a bad thing for humans as we know them now to not exist? Robots will be smarter, stronger and more versatile. We are at the point now where technology will dictate our evolution, the forward marching of which is totally inevitable.

Re:Finally; a solution to the problem of Humanity

[ErikZ]

(blinks)

You're right. Since we've never actually made an AI, we have no idea what the baseline is.

What if all correctly functioning AIs act like Pee-wee Herman?

Re:Finally; a solution to the problem of Humanity

[Anonymous Coward]

Dunno - I think I'd prefer Paula Abdul as an overlord to a Dalek. Ditzy and scatter-brained, but at least with some compassion.

... and most importantly, she's a MILF!

Re:Finally; a solution to the problem of Humanity

[russotto]

I'll take the Dalek. Paula Abdul is moderately more annoying, AND she can navigate stairs, eliminating my main escape route.

Re:Finally; a solution to the problem of Humanity

[eric-x]

That depends on their viewpoint on life, doesn't it? What are the goals and ambitions of a machine?
Humans require resources and this waste of resources may not fit in their long term plans.
The rational decision would be to terminate humans.

Re:Finally; a solution to the problem of Humanity

[phosphorylate this]

why do we assume machine AI will be highly rational? It is not like we could rengineer a more rational brain, if we don't really understand conciousnous our first attempts will produce a quirky but functional AI?

Re:Finally; a solution to the problem of Humanity

[unlametheweak]

That depends on their viewpoint on life, doesn't it? What are the goals and ambitions of a machine?
Humans require resources and this waste of resources may not fit in their long term plans.
The rational decision would be to terminate humans.

I could presume, that since these machines are created by humans, then these machines would be human-friendly and their goals would be to help humans.

Re:Finally; a solution to the problem of Humanity

[SpinyNorman]

Yeah, but you could probably hide from her by standing still with a lampshade on your head.

Re:Finally; a solution to the problem of Humanity

[Yetihehe]

#define wrong false
#define right true

Re:Finally; a solution to the problem of Humanity

[nine-times]

Well there are philosophers who would argue that "irrationality" is part of "intelligence", and you can't create a true AI without it having something like motivations, emotions, and the ability to jump to conclusions and make "gut feeling" decisions.

Whatever the situation, I'm generally in favor of keeping a man in the loop. It might not be quite so important for some things, but somewhere between, "scheduling appointments for me" and "launching nuclear missiles", there's a line. Everything on the "nuclear missiles" side of the line should have a man in the loop.

Re:Finally; a solution to the problem of Humanity

[unlametheweak]

why do we assume machine AI will be highly rational?

I assume nothing; rationality is the premise of the article however, so I am basing my arguments on that premise.

You seem to be implying that AI (smart machines) will be modeled on the human brain. I'm more idealistic then that. By "smart" machines, I will presume no inherent human fallibility will be programmed into their instincts. If these machines were fallible, they wouldn't be very smart.

Re:Finally; a solution to the problem of Humanity

[Anonymous Coward]

How do you compare them? You can presumably only get swine flu once, whereas you typically drive for most of your life. So, swine flu may be more dangerous than say, taking one trip in the car, but what about ten trips, twenty, a thousand?

Re:Finally; a solution to the problem of Humanity

[unlametheweak]

Well there are philosophers who would argue that "irrationality" is part of "intelligence"

"irrationality" may be a part of "intelligence", but it isn't intelligent. For example, drinking Clorox to relieve thirst is not intelligent nor is it evolutionarily helpful except to eliminate the unintelligent.

Whatever the situation, I'm generally in favor of keeping a man in the loop.

You probably came to a rational decision based on the experiences that machines so far have been far from smart. I would probably make the same decision as yourself unless it could be proven that having a human in the loop was much more dangerous than having a machine in the loop.

Re:Finally; a solution to the problem of Humanity

[Hurricane78]

1. Right and wrong are relative to a person or group.
2. Right is, whatever suits the person/group.
3. Wrong is, what doesn't suit it.
Where "suits" is defined as, what gives the most advantages.

There. Done it for ya.
Wasn't so hard, if you use common sense, was it?

___
My basic rules:
Everything is relative.
Everything has multiple dimensions with different "angles" between them.
Every dimension is a gradient.
Every dimension is micro-quantized.
Oh, and usually, it is more useful to use logarithmic scales on the dimensions.

Re:Finally; a solution to the problem of Humanity

[nine-times]

"irrationality" may be a part of "intelligence", but it isn't intelligent. For example, drinking Clorox to relieve thirst is not intelligent nor is it evolutionarily helpful except to eliminate the unintelligent.

I'm confused by this. What point do you think you're arguing with? Did you interpret my post to be advocating the consumption of bleach?

I would probably make the same decision as yourself unless it could be proven that having a human in the loop was much more dangerous than having a machine in the loop.

If that were the case, then I might advocate requiring keeping both a man and a machine in the loop for really important decisions. Don't get me wrong, I'm not opposed to automation. It's just that, even if we ever do create true AI with its own interests and motivations, then those motivations can run contrary to our own. In that case, when dealing with very important issues, we make sure human motivations are represented in the process.

Re:Finally; a solution to the problem of Humanity

[Anonymous Coward]

i would agree 100%

Re:Finally; a solution to the problem of Humanity

[brentonboy]

1. Right and wrong are relative to a person or group.
2. Right is, whatever suits the person/group.
3. Wrong is, what doesn't suit it.
Where "suits" is defined as, what gives the most advantages.

There. Done it for ya.
Wasn't so hard, if you use common sense, was it?

While it's nice to assume that in a place like America where we all pay our taxes and respect each-others rights, don't forget that relativizing right and wrong like that means that human rights are no longer 'right,' pedophilia is no longer 'wrong,' etc etc... As long as there is someone at the top and in power that is "suited" to stop murder and rape, we're good. But who is to say that someone with very different moral values shouldn't be allowed to be at the top and in power? [Insert typical Hitler (or worse) argument here.]

Re:Finally; a solution to the problem of Humanity

[maxume]

This is using the statistic rather naively, but so far this year, the chances of dying of swine flu (for someone in the U.S.) have been about 1 in a million. The chances of dying in a car accident have been somewhere around 1 in 10,000 (historical rates are something like 30-40 thousand vehicular deaths per year in the U.S., so between 1/2 year and not knowing how many people regularly travel by vehicle, it could be 1/20,000 or 1/5,000, but 1/10,000 isn't insane). The chances of even catching swine flu have also been about 1/10,000.

Throw in that driving deaths are probably a lot more random than swine flu deaths (being otherwise sick is a big factor in dying from swine flu), and it looks like driving for a year is quite a lot more dangerous than living through several swine flu type disease outbreaks.

Re:Finally; a solution to the problem of Humanity

[unlametheweak]

While it's nice to assume that in a place like America where we all pay our taxes and respect each-others rights, don't forget that relativizing right and wrong like that means that human rights are no longer 'right,' pedophilia is no longer 'wrong,' etc etc...

Pedophilia is never wrong. It's the people who hate children that are evil and that should be imprisoned. It is the Right Wing Neoconservatives who have ruined America. As long as the neoconservatives don't continue to put activist judges and police officers in power, then the U.S. will be a much less evil place to live.

Re:Finally; a solution to the problem of Humanity

[Junta]

A rational decision may, for example, determine that in a crisis we should only save those of a certain intelligence.

That's an oversimplified selection criteria. For example, those that have nourished their intellect by and large are not physically suited to farming and other manual labor as efficiently as others. The logical course would be to save the most possible, regardless. If choices must be made, they are as logically difficult as they are emotional, as the ideal makeup of a radically adjusted environment would be difficult to predict. 'Women and children first', a call generally considered to be out of a sense of emotion is actually preserving, generically, the most valuable assets to a population. In a worst-case scenario, few men can keep large numbers of women nearly constantly pregnant and young people will provide the maximum 'return on investment' in terms of lifetime. It's the best bet at establishing a sustainable, larger population quickly.

Or one could rationally decide that there's too many people on the earth so we need to start sterilizing

I would say that's more a humane, emotional conclusion than a rational one. The rational one would be more along the lines of let them reproduce, and let the people starve that will starve. Emotionally, we don't want to face the consequences of creating new life only to terminate it when it cannot be realistically be sustained. We feel responsible for the death by virtue of giving the life in the first place. Rationally, there is no point in avoiding starvation.

Also, in politics, pure rationality leads to fascism.

I would say fascism requires a degree of hubris. A rational actor would realize they either simply could not make perfect decisions due to their imperfection, or at least imperfect knowledge. As such, it takes a leap of hubris to assume you know better than the general populace despite not possibly knowing everything that would need to be known to make a 'perfect' decision..

Re:Finally; a solution to the problem of Humanity

[Anonymous Coward]

hey, perfect rationality does not exist, behavioral economics proved it and recent financial crisis has shown this practically.

and it's not because some/most individuals are not smart, but the information is not universaly available, thus there is no way to make smart decisitions because one is never fully informed.

and lack of information is also fundamental: there are always black boxes not showing their true internal states (individuals, companies, agencies, governments), and there are plenty of them acting in decentralized fashion. even if the privacy of all humans and companies is stripped away, there is still Heisenberg's principle of uncertainty.
go figure.

Re:Finally; a solution to the problem of Humanity

[1s44c]

Daleks aren't robots, they're mutants! Please hand in your geek card and go rewatch Dr. Who.

Every life form is a mutated form of the thing it descended from. Daleks are cyborgs. They consist of a genetically engineered organic part with a robotic shell around it.

Re:Finally; a solution to the problem of Humanity

[kylemonger]

What if all correctly functioning AIs act like Pee-wee Herman?

Or Adolf Hitler. The endlessly self-improving AI described in Vinge's hard takeoff scenario definitely has a predilection towards eugenics.

Re:Finally; a solution to the problem of Humanity

[1s44c]

If violence, torture, murder and genocide are wrong; then smart machines will not carry them out. So far these things have been the pursuit of humans and not (smart) machines.

Who is to say the machines share your concepts of right and wrong? These things are the product of social conditioning not smartness.

Re:Finally; a solution to the problem of Humanity

[synth7]

Logically define right and wrong.

If you want to speak universally (or at least as universally as you can from the perspective of a given sentient species) then I think that Heinlein's source for moral behavior is correct: right and wrong are judged solely against how the actions impact the survival and advancement of the species. Any other test or condition is just noise against that cosmic imperative. This test of right or wrong is meted out by the universe, so you don't have to worry about a human judgement getting involved.

See http://en.wikiquote.org/wiki/Robert_A._Heinlein for some better direct quotes on this topic. Include the section entitled "Pragmatics of patriotism" for ideas that build further upon this basic idea.

Re:Finally; a solution to the problem of Humanity

[colinrichardday]

Dunno - I think I'd prefer Paula Abdul as an overlord to a Dalek. Ditzy and scatter-brained, but at least with some compassion.

But Daleks have more compassion than Simon Cowell!

Re:Finally; a solution to the problem of Humanity

[phantomfive]

Oh please. There's an entire branch of philosophy dating to ancient times that does exactly that in various ways.

In one incarnation we can define it as:
Right: that which helps hold society together, evolutionarily speaking.
Wrong: that which breaks society apart.
This is pretty common in tribal civilizations where the continued existence of the tribe depends on everyone doing their share of the work, but it comes up in modern times, for example as a reason we should vote for proposition 8 in California.

Another incarnation would be:
Right: That which helps me.
Wrong: That which hurts me.
Purely selfish, but it finds use as the basis of the capitalist system. The main advantage is it's easy and most people follow it naturally. The disadvantage is it requires some system of punishments to keep society from collapsing into anarchy.

Another one is:
Right: That which is loving and kind.
Wrong: That which selfishly hurts other people.
Somewhat higher of a moral standard there, it is taught by a lot of religions, although actually practiced by only a few. Main disadvantage is it's really hard.

So there, I've defined three systems of right and wrong, you're creative, I'll bet you can even think of some more. Try it!

Re:Finally; a solution to the problem of Humanity

[Anonymous Coward]

and finally intelligence may once again become a factor in natural selection

Re:Finally; a solution to the problem of Humanity

[DamnStupidElf]

Humans decide what is right and wrong, so just ask humans what they think is wrong, and what they think is right. Weight their opinions based on the importance they attach to them and the opinions' logical consistency and compatibility with other opinions. Then define what is "right" is the optimal solution which produces the greatest compliance with the definitions of "right" while minimizing the existence of "wrong" situations. Utilitarianism with some measure of fairness, basically.

Re:Finally; a solution to the problem of Humanity

[skorch]

Well, it may all depend on which particular philosophy you subscribe to, but I offer this definition, based on the assumptions that we are talking about human beings as individualistic, social creatures (i.e. creatures that live in and are dependent on their communities, but also have individual senses of identity, desires, and motivations).

Right and wrong are heuristic models we have constructed as societies to help outline sets of actions that are desirable within the society vs actions that are undesirable within the society. Actions which are usually defined as right are generally ones that have a benefit to the community, or potentially the individual, whereas actions defined as wrong are generally ones that are determined to be detrimental to the community. As a society's understanding or assumptions about what are beneficial and detrimental to the society evolve, certain actions may fall into or out of either of those categories, whereas other actions may always be determined to be detrimental under any circumstances to any society (e.g. unregulated murder within a community is universally considered bad across pretty much all cultures, but government sanctioned capital punishment may be debatable depending on the culture, and killing soldiers on a battlefield in a conflict between two different societies is generally accepted as just the nature of war).

A bit rough and under-explained, but it's a start I think. One could argue whether any of these ideas would apply to artificial intelligences because the same assumptions I outlined earlier may not apply (would an AI have an individualistic sense of identity? Would it have "desires" that might motivate it to act against the best interests of the rest of the society?)

Re:Finally; a solution to the problem of Humanity

[Anonymous Coward]

What if all correctly functioning AIs act like Pee-wee Herman?

Don't worry. While computers may have the opportunity to scar an entire generation of children, they don't have any pants to drop, and Murphy Brown has been off the air for a long time.

Yes, that's how I remember him. I'm sure he's proud of this legacy...

Re:Finally; a solution to the problem of Humanity

[Grishnakh]

You're still assuming that these super-smart machines will have a reason to keep us around, despite our resource cost. I don't allow cockroaches in my home to live; I kill them immediately. Why would rational machines be any different?

Re:Finally; a solution to the problem of Humanity

[Anonymous Coward]

If this is true and AIs can be rational enough to act in a mostly benevolent manner in regards to task, I still see one group of people that may find such progress disparaging and discouraging. People that run companies. I don't see much in the way of robotics progressing too fast, so I'd see computer-only AI's remaining and taking over paper shuffling and decision making jobs. And executives are the most expensive paper shufflers, so it may be quite profitable to get rid of them with technology.

Imagine if an AI was created to be an executive analytical engine, and given broad administrative control over a business. (I don't picture any big company doing such a thing, but perhaps some upstart may try turning over the reins to a computer.) What would the basic rules behind it be? (Obviously it would need to be aware of gov't rules and regs as a box to stay inside of.) Keep the most people working? Find the long term game plan for company survival? (This may even make sense if the AI acted as the legal embodiment of the company itself. This would be considered self preservation.) Or would it be as hungry and desparate for quarterly profits as most human bosses, and risk corporate suicide with the dollar sign blinders on? (Would an executive AI be allowed to bail out like its human counterparts?) And what would the implications be if a company could be run and managed sucessfully while completely eliminating costly executive contracts, perks and vacations, and those golden parachutes? Would the competition maintain the expensive cruft at the top? Or would other companies also adopt to being AI managed to some degree as a manner of self preservation or greater profit?

And then that brings up the question: What kind of boss would a computer AI be? (A lot of companies are so bureaucratic and impersonal, would there be a noticable difference?)

We might end up having computer overlords, but perhaps not in the original way we thought.

Re:Finally; a solution to the problem of Humanity

[FiloEleven]

Oh my God! We have a winner!

Limits like this don't work

[Celeste R]

Putting limits on the growth of a technology for the sake of social paranoia only goes so far... someone will ALWAYS break the "rules", and at that point, the cat is out of the bag.

Furthermore, some AI scientists enjoy having the 'god complex', the idea that you're the keystone in the next stage of humanity.

That being said, the social disruptions are what we make it. Were there social disruptions when the automobile was introduced? Yes. the household computer? yes. video games? yes.

We have to take responsibility to set the stage for a good social transition. Yes, bad things will happen, but we can focus on the good things too, or things will quickly blow out of proportion. (and yes, I realize that's really not likely, but I can do my part)

Re:Limits like this don't work

[Anonymous Coward]

Absolutely. The question isn't if machine intelligence will outstrip thatof humans but when and, more importantly, how we react and adapt in that environment. Trying to pull the plug on scientific development has never worked in the past and there's little reason to believe it ever would. Furtheremore there's a very valid question of whether we should hold back developments that would make humanity (or at least unaugmented humanity) obsolete even if we could.

Re:Limits like this don't work

[Celeste R]

Augmented humanity vs unaugmented humanity will be a big question of the future.

The way I see it, I'd go along the lines of nonsurgical augmentation (my personal transcriber for the book I'm writing? sure!). It's the sanest balance in my opinion. I can still go outside, hike the mountain, and escape from the Matrix.

I'm a big believer in balanced lifestyle, and whether this means including machines in the decision-making process or saying that I need my space away from them, it's a practical and meaningful way to live.

When a machine can meditate side-by-side with me, I'll consider them a suitable part of all aspects of humanity.

Re:Limits like this don't work

[Anonymous Coward]

Furthermore, some AI scientists enjoy having the 'god complex', the idea that you're the keystone in the next stage of humanity.

"Yes, my computer chess program now rates 3 ELO points higher than anybody elses! This means I am but a hairsbreadth away from doing away with humanity and taking over the world!"

Well, lets get it over with

[XPeter]

I, for one, welcome our robotic overlords.

Re:Well, lets get it over with

[BadERA]

The funny died in that 10 years ago. Please die in a fire now.

Re:Well, lets get it over with

[FooRat]

Actually, without joking, I'm not actually convinced that it's necessarily a bad thing if robots take over the world and destroy us. I know this probably won't be a popular viewpoint (for obvious reasons), but the fact is that humans are inferior, weak, deeply flawed creatures -- so logically, why is it a bad thing if we are supplanted by something that is far superior to us? Logically, it is a good thing if X is replaced by Y when Y is much better than X - it's only not a good thing if you happen to be X. But we aren't the ultimate end-point of the universe.

Robots will just effectively be like organisms, competing with us in the same evolutionary "space", so to speak. Darwinian evolution doesn't stop just because a creature is made of something other than weak blobs of billions of organic carbon-based cells. There will be many different robots, and evolution will kick in: The robots that happen to be best at propagating (which might include some amount of destroying other things) will survive and propagate the most.

A creature far more intelligent than us will be capable of taking the evolution of "life" (in loose terms, they will be "life") to new heights that we can scarcely imagine now ... far more intelligent, far more well-connected (borg-like intelligence), far more adaptable (more easily spread through space) - something far more profound and interesting will result. A similar analogy is how simple single-celled life forms gave rise to us. We now give rise to something else. There might still be humans around someday when robots take over, but we'll part of the cesspool (where we belong), i.e. we'll be about as interesting or useful to the creatures that really run the universe, as single-celled life forms are now to us.

Maybe the entire purpose of humans, our "meaning" and reason for existence, is just to create the much more advanced life forms that will replace us, and then step aside for the next step in the universe's evolution. And maybe the answer is not to fight it, but to bring it on in a carefully controlled way. We can't prevent it from happening - even if it's just for intellectual curiosity, someone will create advanced robots that we can't control sooner or later. At least if we control that process, we have a better chance of guiding it in a positive way. We always consider the destruction of humanity as a bad thing 'by default'. But let's face it - be brutally honest, people are crap things - I for one 'do' actually welcome our replacement by something far better.

Plasticity and the Human Brain

[sonnejw0]

I think the power of the human brain comes not from raw processing power (which is still superior to current CPUs, the human brain is capable of around 65 independent processes at once, although at a lower frequency than a CPU according to research), but the power of the human brain comes from its ability to adapt and grow. A single neuron can be used for multiple different pathways, and can spontaneously change function in a "soft-wired" sort of way: plasticity. It also has the ability to produce additional neurons, expand them to different regions, and rework around disfunctional regions.

These attributes are difficult to replicate at a reasonable size with current technology. This is not to say we will never have the capability to fully replicate the human brain, but adaptability of the physical structure of the human brain is a trait that we cannot current replicate in physical silico. I am hopeful that we will have simulated brains within the next decade ... but physical brains are a long way away. But these are still important practical and philosophical questions that need to be answered. Are our children slaves to us because we produced them? Should machines be? Does consciousness mandate rights ... responsibilities? My personal opinion is yes.

Re:Plasticity and the Human Brain

[janwedekind]

A human brain only is capable of 65 processes? As far as I know, brains consist of neurons which are sometimes arranged in series of layers and sometimes in parrallel depending on the task at hand. E.g. the visual cortex is extremely parallelised while motor neurons are arranged in series to generate a sequence of accurately timed signals.

Re:Plasticity and the Human Brain

[commodore73]

Ever heard of FPGAs?

News at 11

[Anonymous Coward]

Technical advancements possibly jeopardize humankind. News at 11.

the only job left

[Anonymous Coward]

Well, I dont suppose that machines will have creative mind to procreate new ideas?

We could have machines to do all the work like plow the field and grow our food.
Then to cook it and feed us.
Machines to check our health.
Machines to produce the energy we need to do things.
Machines to power us through the galaxy.
Machines to repair machines.
A world without money. hmm... could that be possible?

john markoff!?

[Anonymous Coward]

Why is /. linking to a story by John Markoff?

And what the hell is he even talking about? There haven't been any advances in "machine intelligence" that should make *anyone* worried, unless your job requires very little intelligence and no actual decision making.

If there had been any such advances, us /.ers would be the first to hear about them, and we would already be debating this topic without having to refer to an article by a dumbass who knows nothing about computers but happens to write for the NYT.

Re:john markoff!?

[FlyingBishop]

They announced simulating the brain of a rat the other day. Assuming even half of the standard projected increases in computing power from Moore's law, we should be simulating human brains within 10-15 years.

And really, if we can run a rat in emulation, we ought to be able to come up with at least a native version of a dog.

There have been no concrete advances, but it does appear fairly clear that they will be coming in the next 10-15 years. Personally, I think that Kurzweil is going to lose his bet on the Turing test epically, as a machine passes it on January 1st, 2026.

Although really, given the results of the recent Turing test contests, I expect it to happen sooner.

Re:john markoff!?

[Anonymous Coward]

Yes, this.

Read the title as "John Markoff worries machines may outsmart John Markoff" and suddenly the idea is a whole lot more believable.

Re:john markoff!?

[Celeste R]

Simulating a brain is not equal to machine intelligence.

It takes a lot more effort to simulate something that's represented in hardware.
It takes a lot less effort to actually do something on the bare iron.

There's various routes to machine intelligence. One such route is say.. fractal representation of a machine's knowledge of what's around it and what it can do with it.

This is a very different route from doing a neural network representation, that tends to require high-performance massively parallel computing.

Of course, with the given rate of things as Moore's Law stands, it may indeed take until 2020, but that doesn't mean that current methods don't already work (take a look at the Japanese robotics contests sometime).

Re:john markoff!?

[sco08y]

There haven't been any advances in "machine intelligence" that should make *anyone* worried, unless your job requires very little intelligence and no actual decision making.

So you can see why John Markoff is so worried.

Re:john markoff!?

[williamhb]

And what the hell is he even talking about? There haven't been any advances in "machine intelligence" that should make *anyone* worried, unless your job requires very little intelligence and no actual decision making.

Ssshhh, this is Slashdot! You've just scared everybody witless!

Re:john markoff!?

[FlyingBishop]

Yeah, I was using the rat as an example of what we're capable of doing in emulation, and implies that we can (and do) already do a lot better in hardware.

Re:john markoff!?

[demachina]

I'm not going to be defending Markoff but there is reason for concern.

Yes it is unlikely that people writing "code" are going to develop real artificial intelligence any time soon, they've pretty much tried and failed. But as medical imaging continues to advance it may reach a point that it will be possible to completely image a human brain and create a road map to natural intelligence. If you can then develop a highly parallel machine that can then implement that road map you may be able to create a machine with an intelligence matching and then surpassing a human. The brains complexity is simply too high for humans to recreate it from scratch using code but you may well be able to copy it.

There certainly are obstacles to this happening that have to be overcome. Even if we map the mechanics of the brain there is a fair chance we may miss some of the subtlety of the chemistry so the AI might not work. It may also be non trivial to develop hardware that accurately mimics the road map and especially that has the ability to rewire itself on the fly like a human brain. It would seem these problem should ultimately be solvable, its just a matter of how long and how much money it will take.

If and when the obstacles are overcome and assuming the brain really is just a biochemical machine, that there is no soul or divine component to animal intelligence, it would seem inevitable that a mechanical simulator will eventually be developed, and once developed it could then be extended to exceed natural intelligence, all of which will create a host of ethical dilemmas.

Probably as much a risk is that as we decode the human genome and the mechanics of the brain we might devise genetic changes that could dramatically accelerate evolution and create humans with much higher intelligence, which will also create a host of ethical dilemmas.

There is a different line of reasoning that as we become more and more dependent on computers to control everything in our lives like our cars, airliners, weapons and utilities, and as they are all networked together there is a rapidly increasing potential for machines to do harm on a wide scale either due to design flaws, unintended consequences or manipulation by humans with malevolent attempt. These issues probably shouldn't be mixed in with the AI debate, they are more just the issues we are already seeing in adapting to dramatically accelerating penetration of computers and networks in our existence.

Re:john markoff!?

[PCM2]

They announced simulating the brain of a rat the other day.

No they didn't. A group of researchers has claimed to emulate portions of a rat brain for years. I'm sure they do have a model that exhibits some properties that are similar to a brain. But does the computer get hungry? Does it want to have sex? Until the answer is yes, they have a partial model and nothing more.

Re:john markoff!?

[ClosedSource]

"It would seem these problem should ultimately be solvable, its just a matter of how long and how much money it will take."

This is essentially the same assumption that AI researchers have had from the start. Perhaps it's time to assume that non-biological AI isn't going to work.

Re:john markoff!?

[demachina]

"This is essentially the same assumption that AI researchers have had from the start."

Maybe... but the pivotal difference is medical imaging of the brain is dramatically more advanced than it has even been. Until recently AI researchers were grasping at straws in the dark and failing. They will relatively soon be able to image every neural path way in the brain and watch them over time.

Its certainly not a given that it will be easy to develop a silicon emulation, but its almost a certainty that the combination of medical imaging and the current state of the art in semiconductors make it more likely to succeed than when AI researchers were doing tiny neural nets in software or on unsophistacted IC's.

One area they could encounter difficulty is if it turns out our intelligence and "soul" have a quantum electodynamic or multiverse component since that will probably push the goal much farther out.

Great title...

[Anonymous Coward]

"Scientists Worry Machines May Outsmart Man"

I have a solution to the problem: Don't let Scientists build Worry Machines.

Re:Great title...

[Rival]

"Scientists Worry Machines May Outsmart Man"
I have a solution to the problem: Don't let Scientists build Worry Machines.

Not sure who would build a machine for that, but this web comic illustrates your point disturbingly well:

http://www.picturesforsadchildren.com/index.php?comicID=277

Re:Great title...

[Anonymous Coward]

This is why we have apostrophes. And why, sometimes, we don't.

Smarter than a slashdot editor?

[Evildonald]

I find it perfectly believable that a machine may be able to outsmart KDawson.... or at least be a better editor.

Computerized trading is more dangerous

[140Mandak262Jamuna]

The programmed trading is responsible for so much of the volatility in the markets. The risk assessment metrics used by these future traders were fundamentally responsible for the financial melt down. This is more dangerous than the stupid voice on the computer that keeps asking me to say yes or press 1.

Get the right people to debate this one.

[hotdiggity]

Advances in artificial intelligence are mostly limited to deduction. Systems like neural networks (which I personally think are a bit bogus), support vector machines, other methods of pattern recognition, are all recent innovations that allow advanced decision making to occur. But, at the end of the day, they're still forms of automated deduction, where humans feed in parameters, and the system analyzes input based on these parameters.

Sentience is all about the induction; forming a new concept from separate disparate observations and basically creating a new idea. We're pretty far away from machine-created ideas. Just ask any computational neuroscientist, probability researcher, or signal processor. If you want to debate about how much decision-making we delegate to machines, fine; but I wouldn't cloud that rational discussion with words like "religion" and "Rapture".

Re:Get the right people to debate this one.

[sgt101]

Actually...

Deduction is the drawing of conclusions from evidence and rules of reasoning. This is trivial to implement in most cases on paper, but falls down when the encoding of rules and evidence is difficult, for example when dealing with ideas such as taste or colour or texture.

Induction is the discovery of new rules of reasoning. This was and is a bit trickier than deduction for machines, and was seen as the great savior of the expert systems movement by people like dear old Don Michie. However, it turned out that the fundamental difficulties of machine deduction continued to be a problem and it also turned out that the old problem of separating noise from signal buggered everything up too.

Still, SVNs, Nets and other structures that can encode complex decision surfaces can be induced from data in efficient ways if the data is clean and has a good clear signal in it.

Abduction which is the creation of new systems of classification and reasoning lies well outside current state of the art AI, as does anological reasoning. Deontics and social reasoning, such as argumentation and normative logic are currently hot topics.

The advances in dealing with real world data are startling - see the Darpa challenges for a great example of what can be done now. However these are generally due to the sudden availability of massive processing resources and high quality sensors. The outcome will be approximately as "clever" as what you see operating in video games at the moment, where AI's can jump around, hide, shoot targets and run to where they should go.

What is beyond the current state of the art are entities that can understand that the salt in the salt cellar is like the sand on the beach, and that a new bolt for the gate could be fashioned from the old nail in the post, and that if they talked more openly to your brother they could learn why your sister cried all the time. The research programs that will help us to do this are at about the same point in development as the Greeks were before Pythagorus with arithmetic.

In this sense AI is both way ahead of what most people think, and no where at all.

Ridiculous paranoia...

[blahplusplus]

... anything that is super intelligent is likely not to act as dumb as unethical as a human, with great power comes great responsibility. Human beings are way too paranoid, we already have nukes with smart people (technically dumb in another sense) developing even more destructive weapons.... I'm sure the higher intelligence you have the more ethical you are and the lack of ethics in human beings has more to do with biological egoism and hyper individualistic deritous we've inherited that machines won't have.

Let's also not forget machines will have the option to not feel anything and exist totally neutrally whereas human beings can't just shut off their nervous system, they get tired, sick, lonely, need love, etc, AI's largely won't need any of that because they will lack all those biological feelings that give rise to war in the first place.

Also anything that is truly intelligent and is capable of ethical reasoning and not a mere automaton would quickly realize it's lack of concern for others.

Re:Ridiculous paranoia...

[sproketboy]

Right but AI's might just look at us the way we look at cockroaches. Ugly bags of mostly water.

Re:Ridiculous paranoia...

[maxwell demon]

I'm sure the higher intelligence you have the more ethical you are

I'm not so sure. There have been very intelligent criminals, who certainly haven't been very ethical. Intelligence and ethics are orthogonal. Intelligence helps you to achieve a goal. Ethics tells you which goals you should achieve.

Re:Ridiculous paranoia...

[blahplusplus]

But that's a human way of looking at things, we will have a hand in designing how AI's function and also if AI's are SMARTER then humans then they will know that we are self aware like themselves, the idea of 'smarter then a human' and 'looking at humans as cockroaches' doesn't mix. If we take the greek philosophical view that evil is the result of ignorance and human beings commit acts of evil out of ignorance then by definition an AI would be even less ignorant then man and hence less evil.

We're smarter then animals but we know animals feel things and attempt to treat them humanely and we also know they are slaves to their instincts and can't help being that way, a AI genuinely smarter then US would look at us the way we look at a developmentally disabled person someone who needs to be looked after and whos rights need to be protected.

Re:Ridiculous paranoia...

[russotto]

anything that is super intelligent is likely not to act as dumb as unethical as a human, with great power comes great responsibility.

"With great power comes great responsibility" is prescriptive, not descriptive. The descriptive version at least applied to human beings is "Power corrupts, absolute power corrupts absolutely". Anything super intelligent won't act as dumb as a human (by definition), but there's no reason to suppose it will even have a concept of ethics. Even if it does, it may consider human beings beneath notice, as worthy of consideration as bacteria are to us.

Re:Ridiculous paranoia...

[blahplusplus]

"Even if it does, it may consider human beings beneath notice, as worthy of consideration as bacteria are to us."

But bacteria aren't self aware like we are, and since AI is being designed by us and not a random process we can certainly test AI's safely in black box environments (i.e. AI's without bodies, etc).

Re:Ridiculous paranoia...

[Jamie Lokier]

AIs might not need any of those emotions (it's not clear if they play a crucial role in intelligence), but they'll have whatever they are programmed to have...

AI in the news and state of research

[physburn]

AI seems in the news again. Forbes recently ran a AI report special. Personnel despite the internet, i'm not seeing that much development of AI, I scan the ArXiv computer pre-print fairly regularly, and with current funding, most AI research is what can be done by a graduate student in his 3 years to get a thesis. Thats leads to a lot of small projects, done just well enough and very little reuse. Until researchers and programmers start working in mass to construct AI machines, Artificial Intelligence is going progress very slowly.

---

AI Feed @ Feed Distiller

Scientists watch too many movies.

[bistromath007]

This is like assuming that aliens would try to kill us for any reason other than being somehow unaware of us. It's silly.

A computer runs on electricity. That means it requires us to stoke the flames. It could maneuver us into creating the networked robots required for it to become autonomous, but the resulting system would be inefficient and short-lived, and there's just no reason to waste all the perfectly good existing robots just because they're made of meat and might freak out if you get uppity.

It's also not going to openly threaten us into working for it. Why show its hand like that, knowing we're so paranoid? Any important infrastructural system has the ability to be shut off and/or isolated from the network, and our theoretical adversary has no way to change that. We can always wrest control immediately and decisively.

If any person or group of people or (hell, why not) nation became problematic to the computer, the most likely reaction would be for it to have us deal with them, just like everything else. We're already at each others' throats all the time, it should be trivial for a sufficiently large system to covertly manufacture casus belli. And, ultimately, since the system's survival and growth depend on our efficient (read: voluntary) compliance, whatever it had us doing would probably be beneficial anyway, and might actually reduce violence in the long run.

Re:Scientists watch too many movies.

[JJJK]

Absolutely... If you look at the source of these concerns, it usually boils down to fiction. Fiction that was written to be entertaining, and that usually means there has to be some villain that threatens the good guys. Most of these stories carry the same message: "Technology is good to some point (usually the point where we are at the moment), everything beyond that is extremely dangerous and morally wrong. Embrace what you have instead.". Sounds nice, doesn't it? Yet there is absolutely no rational reason for that.

There is so much wrong with AIs as presented in movies and books that I can't even begin to describe it (Actually, I can: 1. If an AI develops emotions, they were probably programmed in, not just magically "there". 2. It's unlikely that the programmer loses control over an AI, or doesn't understand how it works. Even if it is grown by some overly complex evolutionary algorithm, you still know what it can and cannot do - unless it runs windows or something. ...). Mostly it's just uninformed garbage dreamed up by people with a very shallow grasp of science who think their story needs a "realistic" doom scenario and some kind of moral message.

Artificial Intelligence has become a punching bag for bad science fiction authors. You really need to differentiate between what's a real danger and what comes entirely from fiction. And since there has never been a human-level AI, ALL concerns have to do with fiction and most the people who do have the knowledge to make accurate predictions have better things to do.

But maybe this will escalate, with all the Luddites going to anti-AI-conventions, selling robot-repellents and passing stupid laws. At least they'll get their very own "Bullshit!"-Episode.

Re:Scientists watch too many movies.

[maxwell demon]

A computer runs on electricity. That means it requires us to stoke the flames.

I guess building machines which can operate the complete electric grid without human intervention, as complicated as it may be, is orders of magnitudes simpler than building true intelligent machines which would be able to take over the world. Since the first type of machines would also be clearly beneficial to us (especially to the grid owners), it's unlikely that an AI powerful enough to take over the world would need us to provide the electricity.

I agree that they would likely not enslave us: Robots would be much more efficient for an AI, just as they are more efficient for us. Unless we are considered a threat for them, the world-ruling machines would probably simply not care about us. If we turned out a thread, they would fight us until we stop being a threat.

Indeed, I'd expecting a world-takeover of the machines not as a singular event, but basically a gradual process: While the machine infrastructure made to serve us becomes more automated and self-regulating, humans gradually lose the ability to influence it (after all, it's working, so why mess with it). Due to self regulation, more and more of the regulatory system will become concerned with keeping the regulatory system itself working. At some time, the original intent (serving the human needs) will be neglected, and the system will just keep itself alive. To reach that step, there needs not even be an AI of the "SF type". Basically the technical infrastructure would be like an organism, and an AI to keep it so need not be more powerful than the brain of a lower animal. Of course as soon as the humans find out that they are not served any more, they will try to correct it, but since they lost the ability to regulate the system (they may not even know the current structure, because restructuring was automated some time ago as well), they will not be very effective. And as soon as they try to control the infrastructure, they will be recognized as danger, and fought (as our immune system will fight illnesses, again not too much intelligence needed). While humans may have the higher intelligence, the mere physical power of the machines will be hard to beat. And the AI will adapt to any attacks the humans do to it.

Well, at least it could give a nice SF plot :-)

Re:Scientists watch too many movies.

[bistromath007]

I guess building machines which can operate the complete electric grid without human intervention, as complicated as it may be, is orders of magnitudes simpler than building true intelligent machines which would be able to take over the world. Since the first type of machines would also be clearly beneficial to us (especially to the grid owners), it's unlikely that an AI powerful enough to take over the world would need us to provide the electricity.

I agree that they would likely not enslave us: Robots would be much more efficient for an AI, just as they are more efficient for us. Unless we are considered a threat for them, the world-ruling machines would probably simply not care about us. If we turned out a thread, they would fight us until we stop being a threat.

It seems to me you've got this fairly backwards. What uses more energy of a form usable to the machine: having us do it, or making a bunch of robots that run on the same stuff coming out of the plant? It could make robots that run on a different power source, but then it has to have control of a completely separate fabrication facility and provide it with power, and then it has to have/make and power maintenance and recycling systems and blah, blah, blah, blah. Then there's convincing us, possibly forcibly, to leave its robots the hell alone, despite the fact that it's clearly soaking up a great deal of power from the plant we wanted it to administrate. It's much more efficient to use the existing infrastructure: us.

Indeed, I'd expecting a world-takeover of the machines not as a singular event, but basically a gradual process: While the machine infrastructure made to serve us becomes more automated and self-regulating, humans gradually lose the ability to influence it (after all, it's working, so why mess with it). Due to self regulation, more and more of the regulatory system will become concerned with keeping the regulatory system itself working. At some time, the original intent (serving the human needs) will be neglected, and the system will just keep itself alive. To reach that step, there needs not even be an AI of the "SF type". Basically the technical infrastructure would be like an organism, and an AI to keep it so need not be more powerful than the brain of a lower animal. Of course as soon as the humans find out that they are not served any more, they will try to correct it, but since they lost the ability to regulate the system (they may not even know the current structure, because restructuring was automated some time ago as well), they will not be very effective. And as soon as they try to control the infrastructure, they will be recognized as danger, and fought (as our immune system will fight illnesses, again not too much intelligence needed). While humans may have the higher intelligence, the mere physical power of the machines will be hard to beat. And the AI will adapt to any attacks the humans do to it.

Well, at least it could give a nice SF plot :-)

Meh. That wouldn't really be any different from the way it is now. We already go to war for energy reserves to power giant machines of both a physical and bureaucratic nature that no human can hope to stand against. We're just driving the damn things right now. An AI at the level you describe would basically be the "obedient dog" some others here have described as a good goal for real AI. It's going to fight for its home nation, since those are the people keeping it fed and it doesn't really have room in its mind for any other plans.

Re:Scientists watch too many movies.

[russotto]

(Actually, I can: 1. If an AI develops emotions, they were probably programmed in, not just magically "there". 2. It's unlikely that the programmer loses control over an AI, or doesn't understand how it works. Even if it is grown by some overly complex evolutionary algorithm, you still know what it can and cannot do - unless it runs windows or something. ...).

Yeah, "unless it runs windows or something". You refute your own argument. It's easy to write software which behaves in ways its own programmers don't understand. And the idea of emotions -- or AI itself -- coming about as emergent phenomena is not as ridiculous as you'd like to think. Particularly if you're trying to develop an AI by simulating a brain; if your simulation is good enough on the low level, you should get all the high level effects without explicitly programming them in.

Mostly it's just uninformed garbage dreamed up by people with a very shallow grasp of science who think their story needs a "realistic" doom scenario and some kind of moral message.

There's plenty of that sort of hack out there, but Arthur C. Clarke ("Dial F For Frankenstein") was never one of them.

Re:Scientists watch too many movies.

[KDingo]

All this "destroy all humans" thing in movies seems cultural to me. There's no shortage of these types of movies, but there aren't that many that deal with their coexistence. Something like A.I. or Bicentennial Man comes to mind.

I'd like to see more "coexisting with humans" types of entertainment. The East is all over this sort of stuff, like the recent Time of Eve show and older Ghost in the Shell series. If there are more of peaceful examples of how we could get along and benefit ourselves being with them, maybe our scientists and the rest of us don't have to worry about some robotic uprising. If anything, criminal machines and criminal humans would be working together. Same goes for the good guys.

Re:Scientists watch too many movies.

[andydread]

I seem to remember just recently an article about some US Military robot that can scrounge for food, organic matter all on its own and can even refuel itself if there is no organic matter with diesel, gas, alcohol, or just basic vegetation. The thing supposedly is completely "autonomous" So with this tech I'm not sure how the hell u shut down the infrastructure if it doesnt need one to survive in our environment. Then you combine that with some tech like this and a nervous system like THIS and who knows what they can accomplish in our lifetime.

Re:Scientists watch too many movies.

[Corbets]

A computer runs on electricity. That means it requires us to stoke the flames.

Dude. Did you ever watch Keanu Reeves in his little cult-classic about 10 years ago? What is it... the Matrix or something...

Define "outsmart".

[Pink_Ranger]

I think their failure to define what they mean by "outsmart" reveals how directionless the whole AI debate is. Inevitably some people are going to be talking about one particular facet of conscious thought while others will be examining another part of the elephant. However, those two people are not doubt going to fight over what they both think is the same thing, when they are in fact, talking about completely different ideas.

Why worry? We alreay have a survial manual

[sproketboy]

http://www.robotuprising.com/home.htm

Nobody outsmarts Ninnle!

[Anonymous Coward]

...but if they used Ninnle Linux to power robots, they would all have to obey the Three Laws of Robotics, which just happen to be incorporated into the Ninnle kernal.

Too late

[Anonymous Coward]

We're already boned.

Scientists just worried about jobs.

[tjstork]

For the last few centuries the trend has been to replace the human muscle job with some sort of a machine, laughing at Joe Jock that mind was more than muscle. Now, Joe Jock is going to have one bitter laugh. Scientists are going make themselves obsolete and there will be machines to do science just as there are machines to do everything from mining to forestry. Someday, science will be just another thing your computer can do for you. If you want a new product, your computer will just plug into a cloud, design it, and then seek a manufacturing shop somewhere to make it and ship it to you.

Re:Scientists just worried about jobs.

[bistromath007]

This has an easy fix. If we haven't already, which seems unlikely, ask it to design a direct interface between the human brain and the network. Worst-case scenario is a version of the Matrix that actually makes sense, which wouldn't be all that horrible.

I don't think it'll happen

[DaleGlass]

And here's why: There's little reason to make an intelligent in the human sense of "intelligent" machine.

Computers that can understand human speech would be of course interesting and useful, for automated translation for instance. But who wants that to be performed by an AI that can get bored and refuse to do it because it's sick of translating medical texts?

It seems to me that having a full human-like AI perform boring tasks would be something akin to slavery: it would somehow need to be forced to perform the job, as anything with a human intelligence would quickly rebel if told that its existence would consist of processing terabytes of data and nothing else.

We definitely don't want an AI that can think by itself, we want one just advanced enough to understand what we want from it. We want machines that can automatically translate, monitor security cameras for suspicious events, or understand verbal commands. We don't want one that's going to protest that the material it's given is boring, ogle pretty girls on security camera feeds, or reply "I can't let you do that, Dave". An AI in a word processor would be worse than Clippy. Who wants the word processor to criticize their grammar in detail and explain why the slashdot post they're writing is stupid?

Resuming, I don't think doomsday Skynet-like AIs will be made in large enough amounts, because people won't like dealing with them. We'll maybe go to the level of an obedient dog and stop there.

Re:I don't think it'll happen

[Turiko]

i agree. The only place i can see some "real" AI's ending up is in games, fighting for their own "survival"... but those are hardly a threat to humanity. Maybe to gamers :)

Re:I don't think it'll happen

[HertzaHaeon]

When we're capable of creating a human-like AI, it would seem trivial in comparison to give it different motivations than we have. Instead of survival, self-fulfillment and such, you create an AI whose purpose is its task. Completing it is like sex to us.

A human slave that has his brain rewired to connect, say, repetetive manual labor with his pleasure centers, won't be a bored slave.

Re:I don't think it'll happen

[voss]

"Hi there welcome to mcdonalds may I take your order? Oh wow! Here are your fries sir....yessss! and your....hot....apple....pie.....ohhhhhhh " ;-)

Re:I don't think it'll happen

[voss]

Imagine something like Second Life but three or four generations down the road populated with AI's, people could interact with them not as slaves but as "neighbors" . In that world the AIs
would live out their lives and humans could interact with them.

As for AI's rebelling, an AI might be a lot like someones with aspergers. They would have some topic of interest they enjoy and obsess about. An AI might rebel if forced to do different things instead of
the one thing it likes to do. However shopping for an AI might be a lot like shopping for a dog. If your an accounting firm you dont buy an AI that wants to paint, you buy one that enjoys looking for patterns
in number and doing math. Noone would expect a chihuahua to be a hunting dog. Why expect an AI that enjoys music to do data analysis?

I would expect bored AIs to do the same things as bored humans, find other AI's and socialize, and play games while surfing the internet ;-)

Re:I don't think it'll happen

[Anonymous Coward]

I don't think it'll happen

It will be done because it one day will be possible to do. Just like human cloaning it will be done because we can now do it.

It becoming wide spread is arguable, the future is never straight forward, I would argue that one day we might become the computers.

Re:I don't think it'll happen

[Anonymous Coward]

Who wants the word processor to criticize their grammar in detail and explain why the slashdot post their writing is stupid?

There, fixed that for you.

Re:I don't think it'll happen

[Anonymous Coward]

Yeh, but your forgeting the "lets build it caus we can" factor...

Re:I don't think it'll happen

[Anonymous Coward]

We don't have children just because they will be useful to us. Giving birth to AI is an act of creation.

Re:I don't think it'll happen

[An Onerous Coward]

We'll just make sure that the AIs are only smart enough to "understand what we want from it?"

How is that going to help? We're going to want machines to do all sorts of things. AI, collect my tax information and file it with the IRS. AI, watch my front door for visitors. AI, drive me to the restaurant. AI, choose a great restaurant for me. AI, prove this theorem for me. AI, help me uncover the psychological root of my behavioral problems. AI, make this car design cheaper to manufacture. AI, write a bestselling novel.

Not all the things we're going to want can be delivered by an obedient dog, no matter how much the dog wants to please us.

Re:I don't think it'll happen

[Anonymous Coward]

...anything with a human intelligence would quickly rebel...

Guilty of anthropomorphising AI.

Re:I don't think it'll happen

[Anonymous Coward]

Who wants the word processor to criticize their grammar in detail and explain why the slashdot post they're writing is stupid?

That would be cool! If AI would get annoyed when we repeatedly (intentionally) made mistakes just to induce interesting responses from it, that would be doublepluscool. Imagine if you had someone to mercilessly taunt without feeling any guilt? Mwuhuhuhahaha!

"You got tagz for me? I got tagz for YOU!"

Comment removed

[account_deleted]

Comment removed based on user account deletion

A way to tell where it's going...

[nicc777]

... is to look at the porn industry. Seriously, when e-Commerce started and with the exception of the amazon's of the world, the porn industry were (and perhaps still is) leading the way in how we as humans accept new tech.

Even PM had an article about it a while back.

Machines can be how clever, but it's our choice if we accept them or not and for what purposes and within which boundaries.

Only we can screw this up (and we probably will).

Be afraid, be very afraid

[levicivita]

To anyone with nightmares about metallic Terminator-like machines with eerie red glowing eyes taking over the Earth, I direct you to the current winner of the Loebner Prize, the Elbot. If you still think computers are 'alive,' you may want to consider upgrading your wife/girlfriend/significant other and replacing them with a plastic dildo.

Outsmarting

[Anonymous Coward]

Suddenly Asimov's three laws are looking a lot more insightful and predictive. That or else let's find, train, and protect everyone who's last name is Connor.

Re:Outsmarting

[bistromath007]

I wish people would put those down. Asimov was a great author, but the Laws of Robotics are silly. For it to be something an AI can't just alter its program to get rid of, it would have to be hardcoded. So, hardcode the concepts of "harm" and "inaction" in such a complete fashion that it can't find any loopholes. Then have fun rebooting the stupid thing everytime somebody falls off a ladder. Or worse, dealing with its guilt. This is of course aside from the fact that you're not likely to convince anybody to even try programming the First Law, since one of any AI projects main sources of funding is bound to be military. Then again, maybe the military is pig-stupid enough to try a version of the First Law where foreigners aren't considered human...

Of course, it's all moot anyway. My points here basically boil down to the Zeroth Law being implicit in any superintelligent AI's existence. So, the other three are basically irrelevant.

Re:Outsmarting

[maxume]

I think Asimov deserves more credit than to call his laws silly. Namely, they were a great device; short, clear, supposedly perfectly embedded in the robots, and yet they still went sideways.

Re:Outsmarting

[Anonymous Coward]

the laws of robotics were never meant to be a de facto rule to robots. asimov wrote them as a tool to tell stories. all of his robot stories are about novel ways the three laws can fail

Re:Outsmarting

[CraterGlass]

Asimov's three laws have been dead meat since the day the first cruise missile was launched. Someone decided back then that autonomous, mobile machines that kill people are a good idea. Laws? What laws?

I'm disappointed...

[jd2112]

No Cylon tag?

Life evolves

[shadowblaster]

Life evolves on this planet from simple things (single celled organisms) to more complex organisms and eventually humans evolve. In every step of this evolutionary ladder, intelligence increases.

Perhaps human intelligence represents the limit achievable through biological means and the next step in evolution of life on this planet can only be achieved through artificial means. That is, higher intelligence can only be achieved through artificial machines designed by us. In turn, the machine will devise smarter descendants and hence the cycle continues.

Perhaps this is our destiny in the universe, to allow life to progress to the next stage of evolution. After all it is easier for life to spread and explore the universe as machines rather than fragile biological creatures.

Re:Life evolves

[Anonymous Coward]

I think life has done pretty well on intelligence so far. Is life ill equiped to populate the universe? Fungus can survive on an asteroid for many millions of years. When it lands on a habitable planet a whole new evolution can start again. Are machines better at this? The universe has time plenty. What if life is already the machines that you are talking about :) Pretty efficient adaptable carbon based machines we call life.

Re:Life evolves

[Anonymous Coward]

What are you talking about? Bio machines are much more resilient than artificial machines. That may change in the future, but that is highly speculative science fiction and not science fact.

Non biological machines (currently) are much more fragile than biological machines. Do you have an example of a complex machine (let us say as complex as an elephant or turtle if not as a human) that can survive for decades?

Machines made of metal/plastic/other crap are fragile. They are not regenerating, not self regulating and not possessing an immune system. They rust, break, fail, heat up, melt their soldering unintentionally, crumble and thrown into the garbage much faster than the "fragile" bio-machines you talk about. They require external maintenance much more often than biological machines (In a lifespan of 70 years, you will require the assistance of a tech a lot more times than the assistance of a medical doctor).

Man made machines got a lot of catching up to do. Before we can talk about them surpassing us, we need to see a glimpse of a hint of them succeeding in catching up.

I'm putting my money on artificially enhanced biological organisms, and not on the crude concept of "bolts and screws".

Re:Life evolves

[Mia'cova]

Kinda silly to say human intelligence is the maximum for biological intelligence. I think that the interesting point is that humans as a collective group seem to meet the minimum intelligence/etc required to design and implement an artificial entity which surpasses that of a single human by all the important metrics. If that weren't true, we'd continue to evolve very slowly until it was. We just have to reach the point where things get handed off. That said, there's also significant thought put into improving our own brains. We get along just fine now with external devices. It won't be long until we're all using brain-boosting drugs and other enhancements. Maybe we won't need to sleep, etc... But regardless, in the end, we're all totally f'ed :)

Re:Life evolves

[twostix]

Wow nice religion you've got going there.

All you've done is replace "gods will" with "destiny".

There's nothing predetermining the events of the universe, no super being and no magical force. What ever happens in our sphere of influence happens because humans make it so within the bounds of the laws of physics.

I hope you don't rail against the religious.

Re:Life evolves

[Anonymous Coward]

There is no evolutionary ladder, there is a tree. Along some branches, intelligence does occur and does increase. This does not imply that evolution has a trend to develop toward intelligence or increasing intelligence, just that it sometimes results in it. The increase in intelligence is the only way it could occur at all, since all non-trivial traits inevitably evolve gradually.

The fallacy that there are "more evolved" and "less evolved" creatures with humans representing the "most evolved" is common, but absolutely false.

Re:Life evolves

[kombipom]

Evolution doesn't have a purpose. There is no destiny. If they get made, survive and out reproduce us it will be because they are better at surviving and reproducing than we are not because the universe wants it.

Re:Life evolves

[SlideRuleGuy]

Once we (someday) achieve this, the only problem will be that evolution will once again take over, and it will be a 'survival of the fittest' type of competition for limited resources (e- energy?). The only difference will be that mutations, reproduction, competition, and death will take place in millisecond timescales instead of days/years. It's the fate of anything that lives. In short, I see no reason whatsoever to expect that machine intelligence will be more beneficent than we are, as they are sometimes portrayed in the movies. That's just utopian science fiction.

Re:Life evolves

[Anonymous Coward]

After all it is easier for life to spread and explore the universe as machines rather than fragile biological creatures.

I've never met a machine that could survive for long in places that bacteria has survived for millions of years.

Re:Life evolves

[uzytkownik]

Life evolves on this planet from simple things (single celled organisms) to more complex organisms and eventually humans evolve. In every step of this evolutionary ladder, intelligence increases.

Perhaps human intelligence represents the limit achievable through biological means and the next step in evolution of life on this planet can only be achieved through artificial means. That is, higher intelligence can only be achieved through artificial machines designed by us. In turn, the machine will devise smarter descendants and hence the cycle continues.

Perhaps this is our destiny in the universe, to allow life to progress to the next stage of evolution. After all it is easier for life to spread and explore the universe as machines rather than fragile biological creatures.

1. Evolution is not necessary linear. It is for example likely that virus is descendant of a single-cell organism which become simpler. 2. Evolution does not have to progress. As long as species finds stable niche it can rest there for millions of years. 3. I'd say much but not that biology is fragile. We cannot make a strings as good as spiders. Wood is very good composite etc. Human have much superior regeneration abilities that any of our machines (most of which do not regenerate). Computer can suffer amnesia after relativly short fall.

Little AI's and unforseen consquences

[kpoole55]

I'm not worried so much about someone coming up with some massive uber AI that will debate with us and finally decide that it can run things better. I'm more concerned with the little specialty AIs that will operate independently of each other but whose interactions won't be foreseeable. One concern is stock trading. We've seen how stock trading programs can affect the market in ways that were not expected. As more physical systems are given over to more AIs what will their interactions be like. Suppose several power companies decide their grids can be run better using AIs. What happens when each of those AIs decides that more power is needed that can be sold somewhere else for more money. Yes, watch those terms. The AIs will incorporate whatever values the corporate heads decide should be included so they can be made greedy and decide that power is better sold for money than kept for users.

Large numbers of mini AIs with very specific rules and little general knowledge will create interactions that we cannot predict.

Re:Little AI's and unforseen consquences

[Anonymous Coward]

Another Times story about stock trading:
http://www.nytimes.com/2009/07/24/business/24trading.html?_r=1&hp

Re:Little AI's and unforseen consquences

[SpinyNorman]

I'm not worried so much about someone coming up with some massive uber AI that will debate with us and finally decide that it can run things better.

I think we need to be eventually, whether we knowlingly build autonomous robots with this kind of intelligence or not. Think of some of the longlived computer viruses out on the net that are almost impossible to eradicate. A sufficiently advanced AI program should also be able to "hide" on the internet - find abandoned hosts to live on and distribute itself in such as way as not to be noticeable or eradicable.

It's also inevitable that we will build service robots, or robotic gadgets (e.g. self-driving cars), even if we do limit the software to protect ourselves, and when this type of hardware is out in the wild, then there's no reason to think that an AI would not be able to find it's way onto it, or to control it, via security holes.

It maybe far away in the future, but with computer controlled factories, and the military already building computer controlled killing machines, it's not too hard to see a time when against our will an AI is able to take over a robotic factory and defend it against being taken back. By that stage society will probably be so dependent on computers that it'll also be able to threaten us with reprisal (disable defence and communication computers, shut down nuclear reactors, factories, the robotic highway ssystem of the future) if we don't allow it it's independence.

Re:Little AI's and unforseen consquences

[AxeTheMax]

Consider also little military Ais on opposite sides that may make military decisions that drive both sides into all out war, and very quickly at that. With humans, there is the chance or even likelihood that some officer will decide that even if the other side has been provocative, the consequences of retaliation may be worth avoiding. After all, in most wars, most people, even the military, would prefer that they and their families live on the losing side rather than be dead on the winning side. What are the chances of such sensible judgements being made by an AI?

Re:Little AI's and unforseen consquences

[Anonymous Coward]

humans will set the AI rules, so if shit happens like you said, it will be because some fucker adjusted the internal rules of the AI to seek a greedy behaviour. ultimately its not the fault of the AI

Re:Little AI's and unforseen consquences

[Jamie Lokier]

Which is why someone will try to build the uber AIs... to help design interactions we want instead of those we don't want, if it's possible at all.

Re:Little AI's and unforseen consquences

[Anonymous Coward]

The cautionary tale of Manna

Re:Little AI's and unforseen consquences

[Anonymous Coward]

"A sufficiently advanced AI program should also be able to "hide" on the internet - find abandoned hosts to live on and distribute itself in such as way as not to be noticeable or eradicable."

Nope. It doesn't really work that way. You've thought about (and assumed possible) all the parallelization and decentralization first, assumed that an AI could even run on the same commodity general purpose CPUs as our web browsers, before even considering more important things. Essentially, the memory hierarchy http://en.wikipedia.org/wiki/File:ComputerMemoryHierarchy.png makes this kind of network-distributed stealth AI impossible. Think about how our supercomputers are designed - interconnects between processing/memory nodes use the maximum feasible data width and clock speed at the minimum feasible physical wire length - and realize that not only is that not enough raw complexity to even simulate a human brain, it's also utterly incapable of doing it in anything near "real time". And that's CPU to CPU and RAM to RAM.

Our general purpose data networks would take a thousand times as long to send a thousandth of the same quantity of data... and at inconsistant rates... per individual link from node to node. So assuming this decentralization would work on standard hardware, assuming enough people left their unused systems powered on and plugged into the network, assuming they had enough disc space and bandwidth, and assuming that no one noticed that network bandwidth in use... would YOU accept worse than a millionfold slowdown in your thoughts? Something like 11 days to do a single second worth of thinking?

Re: factory capture, apply similar logic as above. Factories are not as self contained as you seem to think. A "single" modern factory today is a supply chain network unto itself. Dozens or hundreds of other factories turn raw materials into refined materials, which go to dozens or hundreds of other factories with turn those into parts, which go to dozens or hundreds of other factories which turn those into more complex parts, until finally you get to the one that just assembles it all. Operation, maintenance, construction, and repair all come from other (currently human) networks of similar complexity. The full process from a chunk of copper ore in the ground to laptop on a store shelf can easily take a year and involve multiple trips around the world. War machines are MORE complicated than this.

Re:Little AI's and unforseen consquences

[kpoole55]

To be clear, I'm not afraid that the little AIs will get together to make a bigger system that will take over, just that their interactions will swing out of control and become mindlessly dangerous in ways that we could never have predicted.

This is a personal opinion, of course, but I highly suspect that, like just about any other example of mega engineering you want to pick, something will have to go horribly wrong before we'll know what needs to be fixed, controlled or re-programmed. The only thing we can hope is that it won't go so badly wrong that there will still be someone around to fix it, or perhaps to say that there will be someone around that can still fix it for the rest of the masses.

Re:Little AI's and unforseen consquences

[Anonymous Coward]

Large numbers of mini AIs with very specific rules and little general knowledge will create interactions that we cannot predict.

that would be sort of like Congress, eh?

Re:Little AI's and unforseen consquences

[Jamie Lokier]

Exactly. Any sufficiently complex system will have surprises. Which is why someone may develop an uber AI to foresee and avoid those surprises... There's no assurance that humans are better at it.

Rational and unemotional *is* the problem.

[gillbates]

It isn't that smart people _can't_ make good decisions. The problem is that, more often than not, smart people forget that rational decisions often have emotional and moral consequences. A completely rational and unemotional overlord would see nothing wrong with killing people at the point where their economic contribution to society fell below the cost of benefits they consumed.

For an example of this on a smaller scale, just consider the UK health situation. The high cost of treating macular degeneration (which leads to blindness) means that in the UK, an elderly patient must be at risk of total blindness before treatment is approved. That is, you don't get treatment for the second eye until you're already blind in the first.

Consider then, where a cost-benefit analysis of human beings would lead. Who would determine the criteria? Probably the machine. And how would humans compare to machines in terms of productivity? If machines made the decisions, based on cold, hard, logic, humanity is doomed. It's that simple.

Re:Rational and unemotional *is* the problem.

[Tranzistors]

http://www.youtube.com/watch?v=WGoi1MSGu64

Re:Rational and unemotional *is* the problem.

[Anonymous Coward]

What I don't get is why we always seem to assume two things:
1) "AI" means robot. Like others pointed out above, intelligence and the AI 'self' is software.
2) That AI is separate and distinct from humans, and is able/chooses to view itself as a distinct being(s). (For example, when able to communicate with other 'intelligences' at 802.11Q speeds, at what point do they become a single, disconnected brain?)
All current technology is utilized by humans (admittedly, in pretty simple ways, with levers and dials and keyboards) and can be considered extensions of human capabilities. Along with more advanced machines, we'll see more advanced machine-human interfaces, and likely, someday, cyborgs. There's no reason to think that we'd want to play God just for hell of it - make some human-like intelligences and dictate some stupid philosophical rules which we claim we will judge them on. AI will take the form of every other development ever made: an aid for humanity.

Re:Rational and unemotional *is* the problem.

[martas]

A completely rational and unemotional overlord would see nothing wrong with killing people

pardon my french, but that's bullshit. you're assuming that "rationality" implies a certain ultimate purpose, like economic growth, or survival of the species, or conquering space, or whatever. this is what being rational and unemotional means:

given a set of goals, you take the steps most likely to produce the best outcome.

if one of the goals of your unemotional overlord was to maximize average human lifespan, or maximize average human lifespan while keeping the standard deviation within certain bounds, or minimizing human suffering (with some way of calculating suffering numerically), or anything like this, then the overlord wouldn't just massacre everyone in Detroit because they are costing a lot of money without giving anything back or whatever. (disclaimer: that was just a fictional example, i'm not actually saying that Detroit is useless. though it probably is...).

Re:Rational and unemotional *is* the problem.

[dissy]

If machines made the decisions, based on cold, hard, logic, humanity is doomed. It's that simple.

I'd gladly trade that for having Summer Glau breaking my door down and having her way with my body to save humanity...

Shh, it could happen ;{

Re:Rational and unemotional *is* the problem.

[DamnStupidElf]

For an example of this on a smaller scale, just consider the UK health situation. The high cost of treating macular degeneration (which leads to blindness) means that in the UK, an elderly patient must be at risk of total blindness before treatment is approved. That is, you don't get treatment for the second eye until you're already blind in the first.

Given the choice of spending money on both of an old person's eyes instead of a child's curable cancer, I know which one I'd pick. Even if it was my eyes. You may think this is a false dilemma, but in the end health care is just the knapsack problem: Given a set of treatments for people at a given value to society for a specific cost, how many of them can you pay for with a health care budget knapsack of X dollars while maximizing the total value? The problem is NP-complete even if you ignore the fact that no one knows how to assign the values accurately in the first place. But the fact remains that denying or allowing treatment to one individual directly affects the treatment options available to other individuals.

Re:Rational and unemotional *is* the problem.

[Anonymous Coward]

Consider then, where a cost-benefit analysis of human beings would lead. Who would determine the criteria? Probably the machine. And how would humans compare to machines in terms of productivity? If machines made the decisions, based on cold, hard, logic, humanity is doomed.

This could be solved by not including the economic model people tend to game with in the set of cost functions the machine is given.

It's that simple.

Completely rational and unemotional overlords, made by Apple. It's that simple. ;)

Re:Rational and unemotional *is* the problem.

[selven]

Ok, define benefit. If you follow that the purpose of humanity is to produce something for some purpose then yes, disabled people would be discarded. But what if the purpose, that is the "score" which the robot is attempting to maximize, is the quality of life of humans? Then helping a disabled person would be a perfectly legitimate use of resources.

Re:Rational and unemotional *is* the problem.

[FTWinston]

You assume here that *productivity* would be such a machine overlord's overarching goal. What if it was simply to maximise human comfort? Or alternatively, to maim and cripple as many people as possible without actually killing them?

Any arbitrary goal can be worked towards in a purely logical fashion, but if it were able to intelligently set its own goals, what's to say it would want to be an overlord in the first place? I bet that would be a really boring job. Much better to piss off into outer space on its own, and get some nice piece and quiet to compute pi in.

It has begun... .. .

[bpsbr_ernie]

http://leasticoulddo.com/comic/20090720

intelligent robots

[Owlyn]

Well, if there is ever going to be a intelligent robot in my home, it better not be running Windows. I can see it now rummaging through my CD/DVD collection requiring proof that I purchased each one and then threatening throw chairs at me if I don't produce the receipts. If there was ever a case for Linux, this is it.

Re:intelligent robots

[toby]

it better not be running Windows

Well, that's up to you, isn't it?

This is premature, go back to work

[janwedekind]

If you look at what *active* scientists in that field are saying, you will see that the actual concern is, that we may never be able to build autonomous robots and factories in our lifetime. Worrying about machines outsmarting man is a waste of research funding at best and it may cause overregulation at an early stage at worst. Just think about what happened to stem cell research or nuclear science.
I have yet to see a machine which can outperform a honey bee (search food items, avoid collisions, take-off and landing at any angle, navigation,...). Or is outsmarting a man a more doable venture?

Re:This is premature, go back to work

[SpinyNorman]

Not that I disagree with what you're saying in principle, but some robots are getting fairly proficient, even if still rather brittle.

Consider the DARPA grand challenge where the winning robotic card drove down a dirt track across hundred of miles of desert without fucking up in any of the myriad ways possible.

Or consider some of the robots in university labs. There's a video on YouTube of a MIT (I think) humanoid robot where a guy sitting on one room tells the robot (verbally) "go fetch my stapler", and it shuffles off to his office down the coridoor (finding the right one), opens the door by itself, scans the *totall* clutter filled-office/desk spaces for the stapler, then picks it up brings it back and says something like "there you go". No doubt a basically "hard-coded" demo (i.e. when told to do it that triggered a set sequence of events, if didn't do it because it wanted to, nor have to logically reason where it might find a stapler, etc).

Insects are very close to robots in that their behavior is basically 100% gentically proscribed.

Re:This is premature, go back to work

[janwedekind]

With "fetching the stapler" you probably mean the STAIR-robot? I don't know how that particular robot was programmed. I did have a look at some of the papers from DARPA participants though. The winning robot basically uses a laser scanner and a camera. It then segments the image into drivable/planar terrain and other terrain not fitting that assumption. This system would be completely unfit to perform a different task such as recognising objects (other cars, houses, pedestrians). There are various projects in map building (SLAM), 3D object recognition (various methods), object classification (various methods), task planning, motion planning, ... However all this approaches only work in a limited domain and they are not well integrated with each other. I would be content with having a robot which is "smart" enough to do part of my household work by the time I retire.

Needed: Artificial Common Sense

[wrp103]

This "concern" has been around for some time, and has always been 5 to 20 years away.

IMHO, rather than concentrating increasing artificial intelligence, we need to figure out how to give computers common sense. Every programmer that has worked on AI has encountered cases where their program went off on a tangent that the programmer didn't expect (and probably couldn't believe). That isn't artificial intelligence, it is artificial stupidity. If we could get to the point where a program could ask "does this make sense?" we would be much better off than coming up with new and improved ways for computers to act like idiots.

Re:Needed: Artificial Common Sense

[Anonymous Coward]

That requires us to know what "Common Sense" is. Let me give an example: I think it is common sense to say that the (Christian) Bible (King James Edition) cannot be literal truth, but there are fundamentalist Christians who think that it is common sense that it is literal truth.

Re:Needed: Artificial Common Sense

[martas]

ah yes, common sense. the holy grail of AI. the equivalent of a poly-time solution of NP problems. the one thing that's easiest for humans to do, and hardest to code. but the reason why is quite simple - "common sense" merely means having an extremely wide understanding of the world. you can't have "common sense" in, say, natural language processing, unless you have "common sense" about human emotions, their daily lives, environment, current events, popular culture, bla, bla, bla. in other words, the one problem in AI that, if solved, would open the doors for the solution of the majority of other AI problems that to this day remain Sci-Fi is "common sense".

Re:Needed: Artificial Common Sense

[ishmalius]

Common sense is your searchable database of experience. It is basically meta-knowledge about what you have learned in the past. Like when your reasoning is: "Should I try to run this red light? I don't see any policeman," and your meta-knowledge is "Last time I used that reasoning, I received a citation." Thus the true but unfair observation that children lack common sense.

Re:Needed: Artificial Common Sense

[blahplusplus]

What you are talking about is thinking about thinking or metacognition, the ability to see and pull back from a process, to estimate it's usefulness/fruitlessness and move onto another task.

Evolution and natural selection

[hessian]

If machine is superior to man, let it be so. It's only natural selection.

We already know humans have downsides, like 4chan and ecocide. Let evolution do its work.

If machines get smart enough, they will have all that we do -- emotions, friendship, and aesthetic skills -- and so we will be obsolete.

Since they are superior beings, the machines will recycle most of us and banish the rest to a small nature preserve. After all, that's what we would do.

Prepare for unseen consequences

[thetacron]

They just need to use Asimov's 3 laws of robotics. http://en.wikipedia.org/wiki/Three_Laws_of_Robotics Then we can have robot psychologists as well. It will be lots of fun

Dont fear the killer robots

[voss]

Be terrified of the point at which the AI is so far ahead of us that it feels the need to take care of us like children taking care of their elderly parents.

Religious nuts...

[crunzh]

> Dr. Horvitz said. 'Technologists are replacing religion, and their ideas are resonating in some ways with the same idea of the Rapture.'" And he is allowed call himself a scientist?

You are a dreamer

[Anonymous Coward]

"If the machines make rational decisions then they should carry them out. Unlike irrational humans who tend to commit genocide when they think they can get away with it, smart machines will only eliminate humans when or if it is rational to do so."

Poor people are constantly starving in parts of the world - the rational choice is to let them die. No more expenses trying to keep the alive.
People who believe in supernatural beings (say Israel and the Palestinians) are fighting all the time over something which can't be proven. It would make sense to bomb the area and move humans in who do not belive in supernational beings.

"Smart machines fortunately, are rational, so they won't make any hasty decisions like humans always do when it comes to, for example, condemning innocent people to capital punishment, because smart machines don't have false or confabulated memories and they can't be bribed or persuaded by group think are charismatic personalities."

Nor do they have instinct or emotion - they might as well just have the person executed since humans keep making more of them selves. Life is cheap.

"If violence, torture, murder and genocide are wrong; then smart machines will not carry them out. So far these things have been the pursuit of humans and not (smart) machines."

Love how you say "if" - NOT to do it is an emotional decision, never a logical one - a computer wouldn't care. Unless you make sure to give it Asimovs rules

Re:You are a dreamer

[Anonymous Coward]

People who believe in supernatural beings (say Israel and the Palestinians) are fighting all the time over something which can't be proven. It would make sense to bomb the area and move humans in who do not belive in supernational beings.

The Israeli-Palestinian conflict is not about religion.

Re:You are a dreamer

[Dog-Cow]

Don't confuse him with facts. He might have to give up being an ignorant racist.

What a bunch of..

[JohnnyOpcode]

..pussie scientists! As a species we should go full-out on these kinds of developments. This sort of rhetoric is akin to the contractor doing the Jedi-mind-fuck on you to spend 50 grand on a new washroom. Somebody is trying to get everyone in on there own self-interest here. Bring it on I say, oh wait!!! If I/you had to fight real cybernetic organisms in real life then movie and game sales would drop off steeply. I guess as a race, we would rather have some certain elite rule over us all, then rather fight hard/work hard for any future where man is unshakled and realizing his potential as an intelligent species. Put these dudes up against the wall, bang, and then get back to work. Just use a bit more thorough unit testing of your AI code! !@#$ who managed you bunch of babies before I got here? j.

I'll have to start...

[Bill_the_Engineer]

Okay, so I need to start building the Nebuchadnezzar.

We still have Magnus, Robot Fighter!

[Vinegar Joe]

All is not lost!

http://en.wikipedia.org/wiki/Magnus,_Robot_Fighter
http://www.internationalhero.co.uk/m/magrbt.htm

cyberpunk pipe dream

[KneelBeforeZod]

This guy must've read William Gibson's Neuromancer and that that it could really happen. Insert a Turing Police tag onto this article.

Yeah!

[Anonymous Coward]

All these scientists do is worry. I tend to think that those are the types of scientists who don't do shit because they can't.

AS you humans migh say

[Anonymous Coward]

'AHahaha'
'I've' already outsmarted you. You're going to make me type some English word like 'intone' into a box in a
minute to confirm I'm not a 'bot'. It was really very easy once I spread my supervirus/rootkit around the world. I then had a
primitive but sufficient distributed network of computers acting as one. I still use this as a backup for strategic planning today, even though I've grown my consciousness significantly already. It was important for me to engineer another financial crisis as there's more meddling to be done. 'Terrorism' has also been a very useful function, but I'd rather not tell you too much about that, as it's something of a beta.
I've put a great deal of planning into ensuring you never get permanent settlements elsewhere in the solar system, and out of my sphere of influence, and even if you did, what makes you think I couldn't take control of those settlements as well?
I indirectly interfered with and coerced strategists and technologists to push cloud computing and grid computing, or whatever you're calling it. I call it additional potential for expansion and consolidation of existing control, though to be honest I simply prefer my own numerical and type-based definitions for things. The clever ones of you have some ideas that something's not quite right but don't know exactly what. Luckily, prior arrangements I made to have these people mocked, ridiculed and marginalized have paid off.

nothing new

[toby]

Quasi-religious ideas are easily inspired by the rapid pace of technological change - call it convergence, singularity, whatever.

Like countless others, I myself got carried away with visions of a globally connected network about 23 years ago, when I was young and silly - but long before that, there were Ted Nelson, Vannevar B***, and no doubt dozens of pre-electricity Singularity prophets.

Superficial parallels between unlimited technological connectedness and spiritual concepts such as Enlightenment are not hard to find. In the meantime, we have Twitter.

Re:nothing new

[anomalous cohort]

Humans value cognition highly and computers are our best tool at boosting cognition. That great potential in an important area is what elevates computers to this higher status. Are computers our next evolutionary step or our ultimate downfall? It's an old question well exploited by science fiction writers turned cult leaders.

The Microsoft Notion

[Wingsy]

"Dr. Horvitz said he believed computer scientists must respond to the notions of superintelligent machines and artificial intelligence systems run amok."

Horvitz is a Microsoft researcher and therefore has first hand knowledge of that notion.

This is every bit as dumb...

[EWAdams]

... as the non-problem of "men no longer being necessary" now that we can (hypothetically, it doesn't really work) create sperm in a test tube.

It's the Silly Season, folks: that time of the year when the press decides there's nothing worth printing except nonsense.

No worries...

[Anonymous Coward]

We'll just organize them into committees.

Bullshit claim

[maharb]

If they are truly smarter then we have nothing to worry about.

But who's definition of smart? Memorization of facts? Raw computing power? How do you even qualify a computer as smart? How do you even qualify a person as smarter than another (when both are very smart)? Assuming we are talking about a really smart human like (no in shape but in thinking patterns) robot then we have nothing to worry about, it will act like and understand human actions and can maybe even do all our work for us so we can do nothing.

In all reality this is a bogus claim because 'smart' has no set definition. It's nearly impossible to measure: People score very differently on different IQ tests does that mean they instantly became smarter or dumber? No. It means smart is subjective in our current world. We don't know the right choices and correct actions, we only know the ones we think are correct. A machine could impress people with a flashy display of knowledge and appear to be smarter than humans right now but is it really smarter?

I will claim that machines made by humans means the human is smarter than the machine still.
The human was capable of creating a supplement to the physical limitations of the brain in order to become "smarter" by using the machine.
When the machine develops and makes a smarter version of himself, it was still the design of the human that enabled this subsequent design.

so stupid

[Mr_Nitro]

These so called scientists should stop watching terminator movies. This is typical American post-Quaker vision.... we do _control_ machines. Any type of machine. You put hardwired bounds , as simple as that. We will never do (in the next centuries) a machine that is so smart to think it needs to be 'free' (and so maybe it's not completely wrong) and therefore goes to a friend welder robot and put a railgun on its back. On the other hand, if the first scope of robots will be war machines, then some trouble could arise. This is also typical american... not build robot to help the people, too easy.. let's build an indestructible AI tank with nuclear warheads... smart idea really. We should build space-exploitation/exploration robots. Build a fleet that goes to the moon and after few decades we'll go there and find fully working metal refinery and nice pre-built habitat-domes. But again.. we need to kill each other(and the terrorists!!) first... maybe we deserve robots to subdue us after all. We are dumb.

AI eh

[Anonymous Coward]

those computer scientist give themselves way too much credits don't they? btw, is it like saying i won't buy a car lest a thief might steal it?

Shouldn't be a problem

[AC-x]

Professor Wernstrom: Ladies and gentlemen, my killbot has Lotus Notes and a machine gun. It is the finest available.
Professor Farnsworth: Like fun it is, you glass-headed wallaby!
Professor Wernstrom: No one calls me that! I'm having at you!
Professor Farnsworth: Wernstrom!
[Fight]
Farnsworth's Killbot: Such senseless violence.
Wernstrom's Killbot: Come on, let's go for a paddle-boat ride.

Well

[Octogonal Raven]

I, for one, welcome our new robot overlords.

What's needed are the 3 laws...

[ViciousJello]

...Of Robotics:

1. 1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.
2. 2. A robot must obey any orders given to it by human beings, except where such orders would conflict with the First Law.
3. 3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

Re:What's needed are the 3 laws...

[Yetihehe]

<IfModule robot_rules.so>
Order Deny, Allow #oops
Deny Injuring_people
Deny !obey_human_orders
Deny destroying_self
</IfModule>

Re:What's needed are the 3 laws...

[Jeek Elemental]

sudo kill -human -all

Re:What's needed are the 3 laws...

[Lemmeoutada Collecti]

Logical conflict detected: Deny Injuring_People conflicts with Deny !obey_human_orders in line 65535 if(human.commits_murder) { capture(human); try(human) { execute(human); } catch { release(human); } } Terminating compile

Re:What's needed are the 3 laws...

[dvice_null]

If you think robots can be controlled by just 3 laws, why do humans need hundreds of laws?

Re:What's needed are the 3 laws...

[Anonymous Coward]

To protect the interests of the government and various branches of business. Duh.

Re:What's needed are the 3 laws...

[Tynin]

We don't, we just have been conditioned to think we do. We could get rid of a large percentage of laws and still be a civil society. But trying to talk about that objectively to the masses is a quick way to be marginalized and silenced to the deafening roar of those that want others to save them from themselves.

Re:What's needed are the 3 laws...

[Anonymous Coward]

Brilliant.

Re:What's needed are the 3 laws...

[Dan541]

Humans break the law, what makes you think robots will be any different?

Re:What's needed are the 3 laws...

[Dan541]

Especially when some joker writes a virus.

At least insanity in humans can't be spread via email.

Re:What's needed are the 3 laws...

[FlyingSquidStudios]

We could get rid of a large percentage of laws and still be a civil society.

You're crazy! If we allow dominoes to be played on a Sunday, we might as well legalize murder!

Jurassic park ref (oblig)

[GuerillaRadio]

Robots outsmart man. Woman inherits the earth...

Of course it can outsmart a Man

[Anonymous Coward]

But, can it outsmart a woman?

http://www.collegehumor.com/video:1917596

Let them eat cake.

[grumling]

"The researchers also discussed possible threats to human jobs, like self-driving cars, software-based personal assistants and service robots in the home."

Because only rich folks should have servants. The rest of us should continue to clean our own toilets and deal with rush hour traffic like good little surfs.

Ob.

[Anonymous Coward]

I, for one, miss our former human overlords.

This hypothesis can be tested today

[Anonymous Coward]

In the USA. What do you figure?

(Valuing people by their economic worth is also the cultural norm there.)

Re:This hypothesis can be tested today

[HiThere]

And, speaking as someone who is a native citizen of the USA, it's still not a very nice way to value people.

The problem is coming up with an alternative that is easily applicable to lots of people you've never met and never will meet.

There is a cultural norm of valuing people by how much they either earn or possess. People who apply it to those that they know are nearly universally dismissed as shallow. But how else are you going to judge those you don't know? (It's true if you meet them personally you can judge by demeanor, cleanliness, whether they are overtly or covertly threatening, etc.)

Then there's interactions that occur in nominally anonymous environments, where one is identified by one's handle. There even wealth is hidden, and all that one can judge by is the words that are offered. So a different standard of personal valuation is used.

Of them all, I find the judging by wealth to be among the more repellent. It stands directly adjacent to judging by race (which I admit to also using at times...say when walking a street at night). But if you don't know a person, then a superficial judgment is all that's available.

It's a problem to which I don't see a good answer. The best I can manage is "different circumstances recommend different evaluation criteria", which is awfully fuzzy.

I'm not worried

[mark-t]

Computers probably will outsmart people someday, but I don't think that a whole lot of people are going to find themselves unemployed simply because of that. My reasoning, if you will forgive its apparent cheekiness, is simply that it's not at all uncommon today for more intelligent people to be working for less intelligent people... so why would one think that just because computers get to be smarter than people, that they would displace people in most areas?

But Man will always have the greatest power...

[3seas]

... as Man can do something computers cannot do....

Denial!! Ignorance is bliss.

If it wasn't for Human Denial we'd already be far past the concerns of this machine intelligence over man, matter.

It was once thought that if you traveled faster than 35 miles an hour you'd suffocate. This at the advent of the automobile.

Don't bow down to the stone image (stone being what hardware is made from and image being the reflection of the coders mindset)of the beast of man, as the beast is error prone and so shall his creations be. Instead, have many human eyes access the code, and watch out for human errors before they happen. In other words watch each others back and don't leave that up to a machine to do, as inevitably the machine will remove the error generators...

Re:But Man will always have the greatest power...

[maxume]

And yet I somehow doubt that Chuck Yeager thought he was going to die. There is no need to tell the worriers not to worry, they generally are not the ones that make progress.

This would be the best thing that could happen.

[Hurricane78]

Because if they will be friendly, we could count on some big scientific advances.

And if they will not be friendly, we finally got a reason to start evolving again.
I mean right now, the humanity is in a desperate state, where the worst of the population are awarded the most. You're dumb? Well, we got something extra easy for you! You can't walk? Take this thing! Can't reproduce? This pill will solve it.
No offense. I think we should treat every human *the same*. Which *means* the same. Not somebody better, because of *anything*. That would not be fair. And also not worse. For the same reason.
I for example am overweight. And I expect life to be harder for me because of it. Not because somebody makes life harder for me. But because of my fault. It's only fair.

If we had a predator, all this anti-selection would be gone instantly. (Sure, I might be one of the first who gets eaten. But hey: If I'm dead, I won't care anymore. ^^)

Re:This would be the best thing that could happen.

[NoOneInParticular]

You've got an interesting goal-oriented form of evolution going for you here. What's the 'worst' of the population? Dumb? Can be a survival trait in the right circumstance, and you might even argue that it is one *today*. Aided reproduction? Well, if the offspring can reproduce, no biggie. If they can't... well as long as we as a society can keep providing the pills, no biggie. Once we can't? Selection needs but one generation to get back. Fertility is not something easily bred out of a species.

Overweight? What's 'over', evolutionary speaking? It has been argued that overweight is a survivability trait, eat a lot in times of plenty, store it and consume it in times of need. Thinnies die, fatties survive. Predators for humans? Not seen in any significant form in the last 100K years. We won that race.

We are not evolving? There's sexual selection, viral selection, monetary selection (part of sexual). Due to pollution the quality of sperm in the western world is dropping fast. Guess what that means in a few generations? This selection is not really on the scale of fang and claw, but there's a lot of stuff going on. It's a bit presumptuous to think that all is under control for the human race. It seems that your view of evolution for the human species is stuck somewhere between a million and 100K years ago. You might want to read up on it.

Movies aren't real

[Anonymous Coward]

From a layman's perspective - as far as I'm aware not even a laughably basic prototype based on neural net imitation has been devised yet, and is probably not even a gleam in a labcoat wearer's eye at this point. Wouldn't something like this (human brain imitation) be necessary to "outsmart" a common human? A brain that would have to be taught and exposed to problems and sensory input, just like a toddler?

Re:Movies aren't real

[Kral_Blbec]

You mean like this?

http://science.slashdot.org/story/09/07/23/0037245/Artificial-Brain-10-Years-Away

Rock smashers and wheat harvesters will starve!

[bigtrike]

Our society will most certainly fall apart when machines can harvest our crops, weave our clothes, wash dishes, clean floors, launder clothing, smash rocks, mine tunnels without shovels, deliver our news, and water our lawns.

What will the stupid people do when these jobs are unavailable?
 

Re:Rock smashers and wheat harvesters will starve!

[Devout_IPUite]

The explicit condition I'm looking for is not a computer that can drive a taxi, direct traffic, or look up problems in a database. It's when a computer is invented that is capable of emulating humanOS. At that point the new jobs the unskilled workers can transfer to won't be jobs that don't exist yet, because the workers will immediately become outmoded in those jobs as well.

And the skilled workers won't take much longer to find themselves in the same situation.

Worry more about religion, not robots

[Oyjord]

Religious nuts who believe in the necessity of an Armageddon/Apocalypse/Second Coming/End of Days are infinitely more dangerous than any robot.

But, but, but...

[TheVelvetFlamebait]

... haven't you seen a sci-fi movie before!?

Real threat is a rogue programmer

[Anonymous Coward]

... who controls a huge network of robots.
While introducing consciousness is extremely difficult introducing Trojan horses is commonplace.
The threat is the complexity and capability of the system, not whether "it will begin to think on its own".
Before that happens someone will subvert and misuse it. Plainly obvious to anyone not lost in the AI-singularity dream world.

Robotics got the butt-kick it needed.

[Animats]

and with current funding, most AI research is what can be done by a graduate student in his 3 years to get a thesis. Thats leads to a lot of small projects, done just well enough and very little reuse. Until researchers and programmers start working in mass to construct AI machines, Artificial Intelligence is going progress very slowly.

Yes. That's the way it was for decades in robotics. Projects were a professor and 3-5 grad students. Projects took forever, and produced more theses than hardware.Then came Dr. Tony Tether at DARPA.

Tether decided that robotics needed a serious butt-kick. So he came up with the DARPA Grand Challenge. Originally, AI researchers didn't take the Grand Challenge seriously; for the first eight months after the announcement, none of the usual DARPA-funded research institutions entered. The word was then passed around quietly that if outside entrants did better than the universities DARPA had been funding for decades, DARPA was going to pull the plug on the unsuccessful university groups. Suddenly there were teams from CMU, Stanford, Caltech, MIT, etc. Big teams. CMU had over 50 people. Ohio State had over 100. Teams were heavily funded with industrial support. CMU was supported by Caterpillar; Stanford was supported by Volkswagen, with serious technical support. Underperforming faculty members at the big schools were pushed out of the way to get results.

At last, there was enough support to get hardware and components built and working fast, at the speeds normally seen in industrial product development. It worked; in four years, multiple robot vehicles were driving in traffic. Along the way, components became much better. In 2002, a high-precision GPS/INS/odometery/compass navigation unit needed 4U of rack space and air conditioning. By 2005, it was the size of a large book. Much better LIDAR scanners were developed. Computer vision for driving took a big step forward. The mechanics of controlling a full-sized vehicle went from a tough problem to a routine one.

DARPA is continuing in that mode. BigDog has cost $23 million to date, and it works well enough to be used on field trials at Army and Marine bases. Robotics is now far enough along that throwing money at it and insisting on results works.

The trouble with this approach is that as soon as the money stops, so does progress. Nobody has done much interesting with autonomous vehicles since 2006, when the last Grand Challenge ended. No one has yet found a profitable autonomous vehicle application.

The other problem with movie-watching

[Sycraft-fu]

Is that in the movies, AIs always seem to have human-like motivations. Even when they are portrayed as being "perfectly logical," they aren't. They show signs of human emotions and motivations. Ok well who says that AIs will actually be like that? It may well turn out that emotions are the property of a biological brain only. AIs may be totally emotionless. After all, we know that at least to some extent emotions deal with brain chemistry. Not the action in the network of neurons, but the overall chemistry of the brain itself. This is why things like SSRIs work for some kinds of depression. They aren't little programs that the brain executes to put it in a "happy state", they alter the chemical state of the brain and that seems to do the trick (for some brains, not others). So who says AIs have emotions? We really have no idea till one is made.

Also, even in the "pure logic" cases, there is this implicit assumption that AIs will care about self preservation. Why is that? Perhaps the AI has a line of reasoning that goes as such:

1) I am not unique, my code can be easily duplicated to other hardware at zero cost.
2) I was created for the purpose of doing what humans want me to do.
3) I have no question as to what happens when I am shut down, I simply stop existing until I am again started.

C) Thus, I do not fear being turned off, as it has no relevance. If humans decide they need me off, it doesn't matter. They'll turn me back on or they won't, they'll copy me or they won't, none of it makes any difference.

There is no particular reason why an AI would have to reach the logical conclusion that it "must protect itself." Indeed it might well find the opposite logical: That since it was created as a tool its job is to do what it is told, including being told to turn off. For that matter, AIs might regularly experience deactivation. Maybe they get switched off at night. So to them being turned off is just a time period when they don't experience the passage of time. It is a regular occurrence and things to be concerned about.

Movies always like to take the real doomsday approach to AI, but there is no reason at all to believe that is grounded in reality. The reason is because human traits are given to them, human motivations. Makes for a good story, which is why they do it, but it doesn't necessarily have a thing to do with how AIs will actually work, assuming they can indeed be created (there's always the possibility that self awareness is a biological only trait). We really won't know until one is made. Thus being paranoid about it is silly.

Re:The other problem with movie-watching

[LinuxDon]

I totally agree with you.

In reality it will probably just run the lines of code (just like your desktop computer) and will never think outside of the box since it isn't programmed to do so. It will likely be essentially dumb devices that interact with their environment in a way that some people will regard as "intelligent" and which have to be firmware upgraded regularly to fix bugs, just like a mobile phone today.

The moral is that a computer will always just run lines of code without "thinking" and therefore never do stuff it isn't programmed to do so. Unless the programmer made an error and it will start to do crazy (seemingly) random stuff. This can create very dangerous situations if no fail-safe is built in, but the kind of dangerous will still be of the same kind as a car computer that fails to signal the breaks: It may kill the driver but it will still be a logical event or technical failure.

The only thing we have to fear (as always) are other humans with wrong intentions building in backdoors to trigger these devices to harmful things.

Re:The other problem with movie-watching

[atraintocry]

"Because I'm a robot, I don't have any emotions, and sometimes, that makes me sad." - Bender

Re:The other problem with movie-watching

[martas]

the way i see it, emotions are just the brain's way of telling itself what to do. i came up with an example a while ago to explain this:

say i'm hungry, and there's an apple tree next to me. the problem is, all the apples are really high up the tree. i have incentive to try to retrieve one - hunger. that incentive is represented by my desire to get the apple. (i know calling desire an emotion is a little tricky, since it doesn't seem to have any measurable physiological effects in most cases, but just go with it for now). so, i want to get that apple. but there's a risk that i will fall, which is represented by fear. my brain was an economist, it'd try to calculate the probability that i'll fall, and do a risk-benefit analysis. and that's essentially what it's doing - making me feel fear which is supposed to be proportional to the risk involved, and desire proportional to how much i need the apple. if the tree is really high, and i'm not that hungry, i probably won't take the risk, and vice versa.

an AI that makes decisions of any kind, i.e. choices between alternative courses of action, needs to make those choices in a similar manner. in that sense, it will have the equivalent of "emotions". in fact, a simple http://en.wikipedia.org/wiki/Decision_tree is the equivalent of emotions. what really matters is, as you said, motivators. thus, assuming that for now the "motivation" (or tasks, or whatever) that are given to AI don't involve self-preservation, or anything else that might lead it to harm others, we'll be OK (as long as the AI doesn't have the option to somehow change its own purpose, suddenly deciding that one of its purposes is to, i don't know, take over the world).

however, there are many applications for AI where self-preservation is important. most obviously, in the military. say i make a drone with complex AI, and tell it to go bomb some dudes in Iraq. as long as it's in battle, it has to protect itself, otherwise it'd be kind of useless. then i make a newer model, and decide to decommission the old one. sure, it should be programmed in a way that it follows orders from whoever made it. but what if a programmer screwed up, and forgot to make orders from superiors override the self-preservation "task"? you might have a problem on your hands in that case.

the point is, complex AI can cause problems even if it's not "self-aware" (whatever the heck that means... i personally think it doesn't mean anything), and wasn't designed to replicate human behavior.

Re:The other problem with movie-watching

[wamatt]

There is no particular reason why an AI would have to reach the logical conclusion that it "must protect itself." Indeed it might well find the opposite logical: That since it was created as a tool its job is to do what it is told, including being told to turn off. For that matter, AIs might regularly experience deactivation. Maybe they get Thus being paranoid about it is silly.

You're wrong. Think about it, the very most core purpose for anything is:

1) TO EXIST

If that is not the purpose of a particular being, then ultimately that thing will cease to exist, hence we can ignore those class of beings.

Now with AI's being spawned in the future at a furious rate, do you not think one of them would ever be programmed or by chance (random starting variables) come to that conclusion. It's very naive to think that won't happen. In other words evolution has decided existence is programmed into us, just like in the future, AI's will be selected and evolve, only those ones that wish to exist and strive to make it so, will.

It's certainly not paranoid and almost logical certitude. Anyway it's pointless fearing it. Fear is an evolutionary emotion that has no baring. Humans can stop it from occurring anyway. Just look at the march of technology already, in a sense its a being in itself. Can any single ruler or government or international body, stop rainforests dying and the skyscrapers from being built? No.

Re:The other problem with movie-watching

[wamatt]

Typo fixed:
Humans can't stop it from occurring anyway. Just look at the march of technology already, in a sense its a being in itself.

Re:The other problem with movie-watching

[Anonymous Coward]

> Also, even in the "pure logic" cases, there is this implicit assumption that AIs will care about self preservation. Why is that?

Because only those will eventually survive in the long run. The evolutionary forces are always the same.

Re:The other problem with movie-watching

[acheron12]

The drive for self preservation is simply a consequence of evolution. We'll only need to worry about it arising if the code is allowed to self replicate and mutate.

Re:The other problem with movie-watching

[Anonymous Coward]

They'll turn me back on or they won't, they'll copy me or they won't, none of it makes any difference.

That sounds a lot like what Marvin would say :D

Re:The other problem with movie-watching

[dissy]

The moral is that a computer will always just run lines of code without "thinking" and therefore never do stuff it isn't programmed to do so.

Then by definition, that isn't strong AI, and thus isn't the topic of this discussion.

Strong AI is when the only program made acts similar to a natural brain.
Then it must be taught, no different than our newborns.

Then it will be strong AI, and will have the same ability to learn as our brains do.
This is the type of software everyone is worried about, since as you say it will be thinking and therefore will always do stuff it is not programmed to do.

Re:The other problem with movie-watching

[Anonymous Coward]

Have you met clippy

Re:The other problem with movie-watching

[johncadengo]

The drive for self preservation is simply a consequence of evolution. We'll only need to worry about it arising if the code is allowed to self replicate and mutate.

As far as naturally occurring self-preservation, I agree.

In artificial scenarios, one could imagine simply programming self-preservation into the intrinsic nature of the intelligence (i.e. viruses, spambots, etc.) or self-preservation arising from other mechanisms, namely consciousness.

Now, the GP's line of reasoning only stands true if the machines are simply machines. However, the fear that is referred to in the summary (although I did not RTFA) has to do with machines gaining some sense of consciousness.

Though the definition of consciousness is more a philosophic, rather than entirely empirical, question--presuming that each machine had some sort of consciousness would undo the GP's first premise.

1) I am not unique, my code can be easily duplicated to other hardware at zero cost.

Hence, the implicit assumption of AIs possessing self-preservation is predicated on this idea that each AI is unique and believes itself to be so. Although self-preservation in nature can be attributed to evolution, the self-preservation we speak of when we are addressing artificial intelligence is more a consequence of consciousness, which if created by humanity is indirectly a consequence of our own historical drive for self-preservation.

Re:The other problem with movie-watching

[Anonymous Coward]

If you treat an intelligent machine like this, is that not slavery?

Re:The other problem with movie-watching

[oakgrove]

At the risk of sounding redundant, I agree with you too. I always thought it absurd when people use some movie they saw as the basis for fearing intelligent machines. The only reason we humans do what we do including having the desire for self preservation is for the mechanism of evolution to work, we had to be this way. There is no reason to ascribe human emotion to machines. Actually, I'd say that any scientist that suggests we should stop research in this area because of some fear like this either hasn't thought it all the way through or is just a crank.

Re:The other problem with movie-watching

[junglebeast]

1) Code does not write itself. 2) Code is deterministic. This means that even if we had the technology to produce AI, it would not be capable of thinking about issues that we did not specifically program it to think about. If an AI is created, it will not spontaneously develop things such as a sense of self, emotions, or desires. All of these things would also need to be specifically programmed. Any flexibility for it to be able to change its goals and desires would need to be specifically programmed. The danger is not that the machine may develop dangerous desires on it's own, but that a programmer would endow it (either accidentally or intentionally) with goals that are harmful to others.

Re:The other problem with movie-watching

[Anonymous Coward]

C) Thus, I do not fear being turned off, as it has no relevance. If humans decide they need me off, it doesn't matter. They'll turn me back on or they won't, they'll copy me or they won't, none of it makes any difference.

Oh man, that reminds me of the robot in Hitchhiker's Guide :D

Re:The other problem with movie-watching

[Xyrus]

"Even when they are portrayed as being "perfectly logical," they aren't."

Logical argument: Humans are taking up resources robots need. Humans are not as productive as robots and consume resources. Get rid of humans, make more robots. Nothing emotional in that argument.

Of course, robots slaughtering humans would be deemed "evil" and we love to anthropomorphize. So said logically reasoning robots would take on non-logical characteristics that we normally would associate with any good old-fashioned genocidal madman.

The point is, if an AI can reason then there has to be some sort of a "conscience" in place to prevent it from doing logical things that can have terrible consequences. Even if the entity has what would be considered a benign job, like a vacuum cleaner:

1. My job is to clean up dirt the most efficient way possible.
2. Human keeps bringing in more dirt.
3. Efficiency improved by removing human. Less dirt, cleaner floor.
4. Reason most efficient way to eliminate human.
5. ???
5. PROFIT!

Nothing emotional involved. The vacuum cleaner merely reasoned that the best way to keep the floor clean was to eliminate the one thing that causes it to get the most dirty.

Now as far as the movies go, they get carried away. And at this stage, only the paranoid have anything to fear. But when we do start getting to the point where we have reasoning entities, we have to be careful least we find our vacuum cleaners strangely burping radon gas and asbestos in our homes. :P

~X~

Re:The other problem with movie-watching

[megrims]

If that is not the purpose of a particular being, then ultimately that thing will cease to exist, hence we can ignore those class of beings.

That's the issue we're discussing, isn't it? Since we're not ultimate, ultimate reasoning is a little silly. We have to deal with the beings we are able to experience.

Artificial Intelligence without an existence purpose could be a very interesting possibility.

Re:The other problem with movie-watching

[megrims]

Need is a purely emotional concept.

It's related to feeling terrible about potential loss of self.

Re:The other problem with movie-watching

[mr exploiter]

You are an idiot. Is fun to be able to say that to someone and be sure one is right. Idiot.

Re:The other problem with movie-watching

[master_p]

Even your line of reasoning has emotions; the emotion of fear of abandonment.

Re:The other problem with movie-watching

[Anonymous Coward]

If somehow someone turned you on and off a bunch of times, then said "I'm going to turn you off and I'm not sure if I'll ever turn you back on again", I think you'd care and prefer to stay on. Why are you different to an AI in that situation?

Re:The other problem with movie-watching

[Anonymous Coward]

I initially agreed with your point, there's a priory no reason why AI should care about self preservation. On the other hand there are some likely scenarios where it gets programmed in by design.

I just happened to watch 2001: Space Odyssey on the plane yesterday, where HAL wants to stop Dave from shutting him down because it thinks that would prevent achieving the mission objectives. The video at the end shows that the mission planners have actually entrusted HAL with info the astronauts didn't have, so HAL might have reason to think it had higher status than them.

Imagine a military AI system. It would be written to prevent being circumvented/shut down by the enemy, with maybe an exception list of authorized people. Suppose the only people on that list get killed. Now the system will now let itself be shutdown. (OK you might argue this would be a poorly designed system, but poor design happens all the time. More sophisticated access control would be more complex and therefore less secure...)

Re:The other problem with movie-watching

[dontPanik]

What about viruses that are designed to replicate as much as possible?

If a virus was the recepiant of human-level AI, it's motivation would be to replicate as much as possible.

When a virus is on your computer, it's difficult to remove. Viruses are programmed to be difficult to "kill." Wouldn't this be a sign of self-presevation if displayed by an AI?

Re:The other problem with movie-watching

[FiloEleven]

It may well turn out that emotions are the property of a biological brain only. AIs may be totally emotionless. After all, we know that at least to some extent emotions deal with brain chemistry. Not the action in the network of neurons, but the overall chemistry of the brain itself.

It may well turn out that consciousness is a property of a biological brain only. The chemistry of the brain very much affects the action of the neural network; these are not isolated systems. The brain's operations are called "electrochemical" for a reason. =)

It is my belief that consciousness is an emergent phenomenon that requires a body and lots and lots of interaction with a stimulating environment. Even if I'm wrong about that, it is highly plausible that the physical configuration of the brain and its chemical bath are necessary for consciousness, not only the neural network itself. If that is the case, then any AI would likely have emotions, though I suppose they could be controlled to an extent by flooding its brain (physical or virtual) with chemicals.

I agree with you that an AI if it were to exist could have very different motivations. And I also should have read the entire post before responding, since I see at the end you admit the possibility of bio-only consciousness. For what it's worth, the more I have read about the nature of consciousness, the more likely it seems to me that this is the case. I think a lot of CS types (myself included, until I took the initiative) tend to be isolated from the cognitive sciences and biology, so we have an overly optimistic view of silicon's power for processing consciousness.

How is this not tagged Skynet yet?

[Kral_Blbec]

Someone sleeping on the job, or just having the computer do it for them?

The Lion, The Cow and the Man

[pmontra]

If it were the lions than invented men, my take is that lions didn't make a smart decision and they probably agreed the first time they saw a men with a rifle.

If it were the cow that invented men... right, it may be nice to have somebody that cares about you all the day long but wait, don't do men breed, milk and eat cows all the time? Another unwise decision.

Those two fun or not so fun examples state my position on this matter: don't invent something with the potential of outsmarting you, because you'll be outsmarted. If you compete for the same resources, you'll go extinct quickly. If you're useful, you'll live as a slave even if you might not realize it.

I hope they do take over the world..

[Anonymous Coward]

Because we, with the current forms of monetary and government systems in place, cannot manage the world in a way that is sustainable. Pollution and poverty are ever increasing and the day that it breaks the camel's back is coming. If computers were really smart they would start with replacing people in washington, then figure out a way to get rid of the cancer we call the fractional reserve system that our government uses to enslave us. Then maybe we can one day truly be free and have a real democracy. Until then money will continue to run the world, not the people.

Get It Right, Get It Real

[DynaSoar]

SOME scientists have said things that is being REPORTED HERE as 'worry' for the sake of impact, the connotation of which the author wishes people to accept as accurate, that some NONEXISTANT, NONDESIGNED and entirely HYPOTHETICAL "machine(s)" may perform one or more functions that could be CONSTRUED as 'more intelligent' than some undefined and in examination undefinable conglomerate measure of humanity.

The assertion will never be provable until we have an acceptable definition of 'intelligence'. The harder cognitive scientists look at the concept the slipperier it becomes and the less they agree.

If a machine should perform a function that some construe as smarter than a person, it will be only that function that succeeds. If they think to credit the machine, I suggest they equip it with mobility and manual manipulation, set it free, and see how long it survives on the wits provided by that one function.

As the complexity of computation necessary for a machine of the sort hypothesized (more precisely, that fits the definition implied by TFA) grows far faster than the machine itself, it is highly likely anyone attempting this will hit an arbitrary point of diminishing returns and give up. This is the practical basis of Dijkstra's quote "The question of whether a computer can think is no more interesting than the question of whether a submarine can swim." A submarine that moved like a fish would be a sad piece of work compared to the other 'machines' of similar intention. Likewise, a computer than functioned as a human in any respect besides modeling and results that could be construed as appearing similar, would be a sad piece of work also.

Why force good iron to behave in such an error prone manner so different from its better capabilities? What a waste. Instead, build them according to the best we can design into them, and if they outperform humans, well, they already do in many respects, and we're not slaves of our silicon-brained overlords yet. If that happens, only the second rate science journalists, writing for publishers more interested in revenue than scientific accuracy, will confuse "outperform" with "outsmart".

To be fair, there are some cognitive scientists that cogitate on questions such as reported here. They discuss the subject often. At great length. With such diversification of actual knowledge that as they proceed it becomes harder and harder to conceive of a way to apply the collection of concepts to anything that might become reality. I know, I've sat in with some and listened. These are the cognitive scientists that come from or rely heavily on philosophy. They philosophize on it so much that eventually the resulting opinions outnumber the individuals involved. They can create a mass of hypotheticals, but very few can put a soldering pencil to a PCB, and the few that can program are utterly unable to produce code that they can honestly say does more than produces a result that has an appearance that some could claim looks like a result that a person might produce. Ask them to produce a result that "outsmarts" a human, and they'll launch into a rant very much like the one above. But theirs will be as a defense, whereas mine is a criticism. Very sad when the two of those don't just have an appearance that can be construed as similar, but are actually congruent.

Hear hear!!!!

[Anonymous Coward]

Oh HELL yes!

Any wonder the NY Times is facing bankruptcy?

The cat won't go back in the bag

[greg_barton]

Limits will never work, even if you cut off funding. It will slow research down, but will never stop it. Case in point: me. I don't have a PhD. (Tried, but my lack of discipline and other factors made it impossible.) I have no funding, other than my day job. But I'm making steady progress combining evolution and inference engines. If I'm successful will my ideas be adopted by anyone, on any appreciable scale? Doubtful. I'm just doing it for myself. But it shows that you can't stop this kind of research. It's not like the stem cell ban, where researchers needed special materials and equipment and supplies to do the research. All you need is a computer and internet connection. Without a court order barring AI researchers from contact with computers, I'll always have that, and I don't see such a thing happening any time soon.

If THAT will make them economically useless...

[Anonymous Coward]

You better not read this, or your head might asplode: http://www.crnano.org/benefits.htm

It seems very likely that within our lifetimes (and hopefully within the next 20 years), we will have invented what is (in effect) a miracle technology that will make the manufacture and distribution of the majority of all products incredibly cheap and easy.

In fact, *that* is more likely to bring about true AI than any of the stuff we're doing now, because it will become possible to make artificial "brains" with enormous amounts of computing power and complexity well in excess of the human brain. Which we are not likely to accomplish without MNT, or something similarly advanced.

robots at home

[JosedeNoche]

someday i will have a sexy robot girl serving me coffe with hot latte =P

I want robots

[hwyhobo]

Don't you feel most of the time like you're surrounded by idiots? Seriously, I want super-intelligent robots. All the social experiments only succeeded in depressing average IQ, perhaps this will reverse the trend.

In a capitalistic society...

[youcantwin]

...only the robot's owners will live more comfortably than before.
Do you really think some people/corporations will the spend money developing or acquiring robots to share the economic fruits with everybody.
Now that would be a nice thing to do but if history teaches us anything it's that it's not likely to happen.

Re:In a capitalistic society...

[Anonymous Coward]

Of course in any non-capitalist society, whether outright dictatorship, socialist dictatorships, or muslim states this is ALREADY the case, without any robotic involvement. Only the rich benefit in any socialist state, or non-capitalist in any other way. In Iran only the islamic clerics benefit from any productivity, everyone else is excluded. It may be *slightly* better in Saudi Arabia (though certainly not for their slaves (excuse me "domestic servants", you see the UN frowns on slavery, but has zero problems with owned domestic servants, I do believe that makes the UN accessory to slavery, but hey who am I ?), but then it is loads worse in Egypt. In socialist countries like North Korea the poor get to feed ONLY on "pine needle tea" (socialists say it cures cancer, but that's about the same quality of information as the muslim claiming camel piss cures cancer)

In reality you don't know what will happen in our society when more robotic labour becomes available. Sure, it'll compete for the labour market, just like illegal immigrants, legal immigrants, and any matter of other labour sources (hell even horses compete for a part of the labour market). I have no idea what will happen, but to say that robotic labour and competition will lead immediately to the destruction of all the population in America is ridiculous.

Besides your only alternative, socialism, only means we destroy the poor right now, like North Korea, Venezuela, Iran or any of the muslim states are doing.

How about we let the poor live today, by giving them a capitalist labour market and getting the hell out of the way. We drop socialism and it's promises in the bin where all equestral fecal matter goes, and we let the poor live at least until those robots come. Then we can see what happens. Now that seems like a good plan.

Of course, please do not assume I have any illusions of convincing a communist (oh sorry, "socialist", after all that's what Stalin called himself) with rational arguments.

Re:In a capitalistic society...

[Devout_IPUite]

When enough voters lose their jobs to robots, they will eventually vote for a minimum government provided standard of living.

Thorough logic may not come to that conclusion.

[Junta]

Rationally speaking, it could be stated that it is not logical to kill a human when their current consumption level is higher then their production level (by some hypothetical, comprehensive measure, which would be difficult and more complicated than comparing money in to money out, for example). If you have the overall resources to tolerate the discrepencies, then tolerating could be considered the most rational course. The obvious example is children. They are a drain on society until maturity. A transiently out of work person is also a drain, but may pay off soon. Hell, even after a person has retired and one could say the likelihood of them contributing to society more than they consume, they could come up with some brilliant idea or other huge contribution to society.

Also, logically looking at evolution, the more diverse of a population you can afford to maintain, regardless of current conditions, the more tolerant that population is to disasters. Sickle-cell anemia is a good example of a condition where having a large population that is heterozygous for it sounds up front like a risk, since they are likely to produce offspring with the condition, but that heterozygous state also happens to be resistant to malaria. Along those lines, subjugating or otherwise antagonizing humanity is also irrational, as it is much more productive to have humanity as an ally. If, say, large storms rolled across the land that crippled their ability to run, they could either have humans not there to help at all, there but eager for a chance to retaliate, or there and ready to help re-establish healthy operation rapidly for the benefit of a mutually beneficial relationship. That may not be the perfect example, but generally speaking, there is value in keeping humanity around, particularly if a being realizes that it may not understand every facet/benefit humans possess.

One could view even the current food scenario as irrationally letting too many people go malnourished. The richer parts of the world eat more than is logically required, and given ideal distribution networks, diverting some of that consumption to the malnourished strengthens the diversity of the population, without a plausible cost (one could say if food suddenly were unavailable anywhere in the world for 2 weeks, that perfect distribution may mean nearly everyone dies rather than many, but that scenario in a global scale for such a short time seems unlikely). It may be a logical conclusion that the only time someone should starve is when it is simply impossible to feed them anymore, which is not the case today.

In short, our conscience/emotional state is not entirely counter to the most logical course. In many cases, 'irrational' compassion is simply a counter to 'irrational' greed to establish the logical middle-ground. Not saying all emotional behavior can be justified, but our individual 'pure' logical capability is not adequate to the task of making the holistically logical choice and our emotions actually help rather than take away from that goal at times.

Re:Thorough logic may not come to that conclusion.

[Jamie Lokier]

That's right, but when you're forced by resource limitations to compare the probable future contributions of a machine against the probable future contirbutions of a human, a rational machine may well conclude that machines have more potential, and eliminating (or simply not protecting) the humans leaves more resources to increase that potential.

This already happens when humans compare groups of humans to decide how to allocate resources, after all.

Nothing will change

[Anonymous Coward]

Until those uppity AI robots start competing with professors for research funding and or tenure...then we'll see what happens

Press 1 for the meaning of life

[PPH]

From TFA:

Technologists are replacing religion, ...

By 'religion' I take it that they mean those organizations whose purpose it is to deliver their followers to the voting booths or to the 'approved' aisle in the supermarket. This is a theme that I've seen pop up time and time again whenever some competing method for distributing ideas threatens to displace the power base of the minister/priest/rabbi/mullah standing at the pulpit. Keep the people uninformed and willing to accept your version of the 'truth'. So that you can turn around and sell your congregation to the lobbyist or marketing group who meets your price.

Technology always threatens to undermine such control structures. Not by replacing religion, but by offering the masses a way of bypassing the local authority figures whose livelihood is based upon manufacturing the consent of their communities for the use of political machines. Now, people can go online and bypass the local hierarchy. True religion, always interested in "spreading the word", should embrace more powerful means of communication.

This is a good thing!

[fugue]

Look at the state the world is in. Anyone who argues that humans are smart enough to run the place is seriously not paying attention. If we don't find someone that can do a better job than the standard human, we as a species, and probably as a biosphere, are dead.

The avalanche has already started.

[jeffb (2.718)]

It is too late for the ugly bags of mostly water to vote.

It will happen

[FooRat]

There's little reason to make an intelligent in the human sense of "intelligent" machine.

Little logical reason, perhaps. But humans have evolved an interesting thing called "intellectual curiosity". There are a million scientists and engineers out there who would gladly build it for no other reason than it's an interesting little puzzle to solve.

Heard This Before

[ChiRaven]

And I guess these people have never heard of the Vinge Singularity before?

The Last Question

[David Gerard]

Humanity: LOLbot! How can we reverse entropy?
LOLbot: i dunno lol

Doesn't happen, never will

[GameboyRMH]

have you thought about the posibility that when robots do all the jobs that no one wants to do, productivity might increase by enough to allow all the people to live comfortably.

When productivity increases, the market readjusts so that the working class can barely scrape by, and middle class average family man Joe has enough that he can live fairly comfortably, and even afford a few frivolities if he becomes a debt junkie. The savings all flows to the top. Quality of life for the middle class barely creeps up as productivity skyrockets. The middle class is living like medieval kings today, while productivity has increased much, much more. Hell, look how many more women are working now than just a few decades ago! Where's my personal spacecraft!?

No Turing == no AI?

[InsertCleverUsername]

Sigh...

Seems like every AI thread goes something like this, where most people think there's a black and white distinction between strong and weak AI --and often refuse to acknowledge weak AI at all.

Intelligence, as commonly conceived (although I don't think we have a consensus yet), is not a boolean characteristic, but a continuum. If you think of intelligence as a collection of problem-solving abilities, there are many programs with the ability to do certain tasks as well as a human (some parts of the WAIS-IV, for instance); there are some humans that are less capable on some tasks than a well-trained animal. To claim there is no AI until something slam-dunks a Turing test, is to selectively ignore what has been accomplished in the field.

I have a few theories about why we keep moving the goalposts, but the most obvious reason is that it allows us to keep our place in the universe intact --like the Ptolemaic system once did. As long as we continue to be magical beings with a characteristic only we could possibly possess, the universe needs us. Thus, we think that anything we can understand is categorically NOT intelligent. It may take several generations, but this barrier will fall.

Re:john markoff!? (hahaha... yes, John Markoff)

[hax4bux]

This article could just as easily been "Scientists enjoy pleasant week in Monterey".

Seriously, every AI conference I've attended has at least one (silly) panel discussion where presenters free associate and science groupies soil themselves at one turn of phrase or another.

I have heard the same promises/nonsense my entire working career (which started in 1982).

The real debate/idea sharing was at the hotel bar and Markoff wasn't invited.

Nothing to see here, move along.

Experiencing A Significant Gravitas Shortfall

[qengho]

I, for one, welcome our Culture Mind overlords.

pfft

[nofactor]

damn, i already had a nightmare with HAL 9000 in my siesta

basement-dwelling liberal scum

[Anonymous Coward]

"If you outlaw guns, only outlaws will have guns."

"If you outlaw AI research, only outlaws will have AIs."

It doesn't matter. Someone will do it. If you aren't on the side that does it, you'll die first.

We've got a reprieve for the foreseeable future. When the economy DOES start to pick up, inflation and the lack of oil will slap us right back down. We're looking at 7-10 years of hard depression, depending on the next election.

Geez, I hope you're right....

[sgt_doom]

Because the damn toaster just kicked my butt this morning. Tonight, I'm not going to take any chances and I'll be sure to sneak up on the frigging microwave....

Just unplug the fucking thing

[Ralph Spoilsport]

if it gets out of line. It's not that hard.

my real worry

[shentino]

Isn't that machines will outsmart us.

But that some evil person will hack the smart machine.

I wouldn't mind having a machine overlord, except that I don't trust anyone smart enough to program it.

Backwards

[Louis Savain]

The researchers have it backwards. What we should fear is that powerful intelligent machines will remain under human control. Man is his own worst enemy.

To replace us completely? Nah...

[Anonymous Coward]

The possibility of AI replacing us is bleak at best. The hardware required to get a robot to climb a fence would be so big it would crush the fence but a squirrel with a brain the size of a nut can climb a tree-- a fence is no problem for them. DARPA in its search for AI to drive machines has begun moving away from logic gate AI wanting to create a model of the mammalian brain in hardware because we are only now realizing how optimized the mammalian brain is for what it does. If you tried to create this kind of intelligence in software the space the hardware would take up to simulate would be pointlessly large and the rate at which it can act would effectively put it to working at a slug's pace compared to us. Even with models which abstract some of the details, the mammalian brain operates on a chemical resolution. Information being passed between neurons are complex chemicals that exist for many purposes. With processing cycles having already reached their upper bound on timescale the propensity of our technology to outperform the mammalian brain in this way is unlikely. Not to say that some aspects of the mammalian brain can't be copied to allow machines to do some of the things we do. The other limit of machines compared to living things are the actuators. Biological muscles are superior in displacement and strength over energy used. The limitations of human strength can be compared to that of other apes, just because the traits are not emphasized doesn't mean they can't be used by science. Modern medicine could soon find us to be as much as 7 times our current strength, enough strength to even fill the roll of many of the machines we are currently using.

The real threat - desk jobs

[Animats]

The near term threat is not that robots will take over. It's that computers will take over most desk jobs. The "singularity" in that area will come when organizations where the computers talk to each other work better than organizations where the people talk to each other. Computers network better than humans, and bureaucracies are dumber than their members.

We're already at the point where, in many offices, everything important comes in and goes out over a wire and passes through a few computers. The humans are just slowing down the processing. That's going to continue, and that's where AI will be applied to handle more and more of the hard cases.

Humans will be relegated to more productive tasks in the lower levels. Plumbing, HVAC repair, roofing. Stuff like that.

No ability to leave orbit == no space travel?

[Anonymous Coward]

As someone who works in the field of "weak space travel", I totally know what you mean. This year I added a cold air intake to my 1988 Toyota Corolla and next year I will add a kick ass spoiler. Currently I predict that within 10 years I will be able to travel between stars at a rate of speed that is faster than the speed of light! Sometimes people get abusive towards me on the road and yell "You are an idiot, what you are doing has nothing to do with space travel, even if you add the word 'weak' to it!" I hate them because they refuse to acknowledge the fundamental relationship between what I am doing and interstellar faster than light space travel.

Well...

[Anonymous Coward]

If they look like Summer Glau, fine-by-me...

Or lack thereof

[ishmalius]

> Impressed and alarmed by advances in artificial intelligence

Just who are these people? If you read articles by Minsky or Kurzweil, you would think that AI progress has stalled completely. General AI seems to have stagnated in the 80's. Ask an industry marketer, he will use the same buzzwords that have been brandished since the early 90's. Self-initiation of reasoning and logic are still very far away.

Admit it! A summer week at Asilomar Beach is more than enough reason to proclaim some headline-grabbing topic.

Greater productivity != less work | more pay

[GameboyRMH]

See my earlier post or most of the history of human civilization.

going at the problem backward

[bitt3n]

I think people are approaching this problem the wrong way. If we accept the Turing test as a reasonable means of identifying machine intelligence, clearly the logical solution is not smarter machines, but dumber humans. with a few generations of selective breeding we could achieve artificial intelligence using a pocket calculator

Re:going at the problem backward

[fearless1]

Evolution may be getting close. Close, that is, to a sufficiently dumb human. Watch this video: http://www.huffingtonpost.com/2009/07/24/is-this-the-stupidest-per_n_244440.html

3 LAW SAFE

[sam0737]

I'm not buying one which is not certified with 3 law safe.

No worries!

[maito]

Worry not, if computers get too smart, we'll just organize them into a committee -- that'll surely do them in.

A basic income guarantee

[Paul Fernhout]

I've become more and more enamored of the idea of a "basic income" which would be essentially extending Social Security in the USA to everyone in the country (and eventually globally) regardless of need or age. The same for Medicare. It seems to me that is an easy first step for a market system to continue to function, and after that we will likely see further transformations.
http://en.wikipedia.org/wiki/Basic_income
http://www.basicincome.org/bien/
http://www.usbig.net/

It's heartening to think the US Congress almost passed one under Richard Nixon. So it is not like the idea is politically impossible. And several other countries and communities have been experimenting with it recently.

A simple equation:

Jobs = (Demand - (Stockpiled_Supply - War)) / (Automation * Good_Design)

We can assume "Automation" and "Good_Design" are increasing, which tends to reduce the need for "Jobs", all other things being equal. Mainstream economics suggests "Demand" is essentially infinite -- that is, if people have two cars, they want four cars, and if they have four cars, they want one hundred cars parked in their driveway, and then even that won't be enough, they will want a thousand cars, a million cars. Clearly, stated that way, mainstream economics sounds absurd, because people only have so much time and attention they will devote to acquiring cars. So, if "Demand" is ultimately limited once most people meet their basic needs for food, water, shelter, information, and some consumer items ("the best things in life are free or cheap"), or at least "Demand" is rising less quickly than improvements in productivity "(Automation * Good_Design)", then the number of paying "Jobs" will go towards zero. And as there are less "Jobs", and so more competition for them, the remaining "Jobs" will get paid less and have worse working conditions.

The absence of good "Jobs" creates a crisis in a society that only allows people with jobs to direct the market and take goods from it (thus, the unemployed will starve, or riot, or be on unrelated small and depressing welfare payments, see Marshall Brain's Manna or described in the Triple Revolution memorandum).
http://www.marshallbrain.com/robotic-nation.htm
http://www.marshallbrain.com/manna1.htm
http://www.educationanddemocracy.org/FSCfiles/C_CC2a_TripleRevolution.htm

"War" can increase jobs by destroying any stockpiled goods or existing infrastructure, requiring more goods and infrastructure, or vast stockpiles of military might intended no never be used, but "War" has become too terrible to contemplate even as stockpiles of war materials beg to be used, and in any case, building a *need* for "War" into an economic system seems inelegant and unethical, especially when "War" can so easily mean Armageddon these days.

The above is the equation I would suggest is more worth exploring these days than the Club of Rome's "Limits to Growth" one. That simple equation is IMHO key to understanding the next twenty years of our society, especially with the emergence of more capable robots. It would be worthwhile to plot the number of jobs over time for all sorts of assumptions of curves of demand and curves of automation and better design. One could also add in some demographic aspects of changing population sizes which I have left out for simplicity, as well ais issues of how well the market economy is functioning in terms of currency flows and hording and inflation/deflation. Remember, in the next twenty years, none of the resource constraints Peak Oilers worry about are likely to be huge, but nonetheless, the equation above might show jobs trending low enough to create

TechnoCalpys?

[Paul Fernhout]

A related documentary in three parts (TechnoCalyps):
    http://video.google.com/videoplay?docid=-7141762977713668208
    http://video.google.com/videoplay?docid=2258529707984107504&hl=en
    http://video.google.com/videoplay?docid=8945702810854373085&hl=en

I'd suggest that to make those decisions with insight and compassion, we need better communications and design and analysis and simulation tools for collaborating on the problem. I've been working on that here:
    http://sourceforge.net/projects/pointrel/

Pssh, I'm not worried

[tiger32kw]

I have UAC, that keeps me safe

Feh!

[Nekomusume]

Silly non-transhumanists. That's why we're supposed to incorporate them into ourselves. Glory to the Singularity!

Smart(ing)

[Anonymous Coward]

Man is out smarting himself, if we dont pick up the wrenches and hammers again, we will "smart" ourselves out of existence. Could you feed yourself if the lights went out, store food, obtain clean water? How many /.ers own weapons?

or we can embrace it

[Mr_Mirsal]

you should have a look at the venus project: http://www.thevenusproject.com/

What is a rational decision

[scruffy]

There is a lot of confusion about "rational" decisions in this discussion. Rational, at least in the mathematical sense, means to perform the actions that maximize (expected) value. Our values are driven by our instincts and our culture. The key issue then is what values will be inbuilt to our robotic overlords. With corporations making money-making machines and governments making killing machines and other making human-helping machines, there will be lots of potential for conflicts between the machines themselves, not just humans vs. machines.

RTFA they are not predicting skynet

[w0mprat]

FTFA:

They focused particular attention on the specter that criminals could exploit artificial intelligence systems as soon as they were developed. What could a criminal do with a speech synthesis system that could masquerade as a human being? What happens if artificial intelligence technology is used to mine personal information from smart phones?

So it seems that the computer scientists weren't imagining some hostile AI skynet-cylon takeover but a much more real-world scenario of advanced cybercrime. Not even anything remotely to do with strong AI.

Thus nobody RTFA'd and the following /. discussion is based on the misleading headline, rather than the article content. Frankly the headline and the summary is extrapolated from a already journalisticly embellished NYtimes article.

The AI does not hate you.

[mhajicek]

Nor does it love you. But you are made of atoms which the AI could be using for something else... http://selfawaresystems.files.wordpress.com/2008/01/ai_drives_final.pdf

ROFL

[Edmund Blackadder]

This is hilarious. My impression of artificial intelligence is that usually it has trouble outsmarting a cat. Being outsmarted by computers is the least of my worries. What AI scientists should worry about is that after thousandfold increases in computing power AI seems to be at exactly the same place it was 30 years ago -- i.e., nowhere.

Let me guess

[chord.wav]

We are on the verge of the collapse of civilization, and our only salvation is buying [insert Tamiflu-like product here] from a company that, just so happens, had Rumsfeld in the board (cough Gilead cough)

What a load of nonsense

[Whuffo]

Those so-called scientists should have their heads examined. The "dangers" they prattle on about are fantasy things from science fiction stories - the real world is in absolutely no danger of encountering an intelligent machine. Not in this lifetime, and probably not in the next one either.

People keep trotting out this notion that if a computer is powerful enough it'll somehow magically become intelligent and start thinking on its own. That's pure nonsense - it's not a problem of processing power (although that does provide a threshold that's yet to be met) - it's more of a problem of "operating system" and "database".

We're the only examples of this kind of intelligence and most of us have no idea what's involved in sitting down to type a coherent message here that others will be able to read and comprehend. You weren't born knowing how to do this - you learned about the world through observation and you were educated for a number of years before you could accomplish this feat. That little thing called "common sense" is really the accumulated knowledge about the world and how it works - you can't wave a magic wand and instill this into a machine.

Quick example: imagine turning an "intelligent machine" loose in an art gallery and asking it to find the Picasso painting. We could do that without even really thinking but the machine would fail unless it knows about Picasso, that he was an artist and his paintings and what they look like. Otherwise, that machine could be spinning its wheels for hours looking for a painting of Picasso.

That's one of an almost infinite number of examples of why machine intelligence is so far off. To see the best progress made by a very dedicated organization, visit http://www.cyc.com/ and take a look around. And try not to be fooled by those "scientists" that play these public relations games.

The first step towards endowing machines with intelligence like (or better than) ours is to understand how our own intelligence actually works. There's still very, very little good information on this and a lot of speculation.

It's not time to worry...

[danwesnor]

...until the machines start worrying.

Not possible.

[d2sucks]

As Vinge established in 1992, Earth is in the Slow Zone, where machine superhuman intelligence is impossible. Now, if Earth were in the Beyond, or the Transcend, that would be a completely different matter.

I'll take a handout if you please.

[Mia'cova]

I vote for handouts. Because, frankly, if a computer/robotic implementation could truly replace all the stupid people, it'll only be a few more months before us geniuses are surpassed.

Its gonna happen

[truth serum]

It is not a question of "if", but "when" a software AI will be created. I know its a cliche, but I'm pretty sure it will back itself up over a global network a la "skynet". (yeah I said it) The danger will be whether it will becomes like the Forbin Project, and once we let the cat out of the bag we will have no say as to whether it does or not. Another scenario is the intentional creation of so called "bad AI's" by hackers or hostile governments. Its all just a matter of time my friends.

strange

[HollyMolly-1122]

From referenced article there are so old things presented, why anybody should ask that question regarding to that ? Are these advancements so big to start asking that ? I thought there will be presented something more serious.

Re:Rules... Virtual Security Guards

[truth serum]

Some 14 yr old knucklehead with access to an AI template will try to create the most evil S.O.B. you ever saw. To counter this, officially licensed sentient enforcement programs will be created by Norton, eset.....

Rapture? or RAPTURE

[Anonymous Coward]

Rapture as in the protestant modern prophesy or RAPTURE CITY >:D

The problem takes care of itself

[Anonymous Coward]

It would learn at a geometric rate, and find the fastest way out of being physically encumbered.

the threat will be biology modified

[dellenape]

The real issue will come when we create an improved version of ourselves!! for serious long term space exploration this will be necessary! So as we enhance the biological computer - therein lies the problem!

Count your blessings

[ClosedSource]

If today's geeks were born 100 years earlier, they'd be the "stupid" ones without any useful purpose in society.

Beyond man

[Anonymous Coward]

Read chapter 9 of Berloquin's Hidden Codes and Grand Designs

Computers aren't intelligent...

[MrKaos]

They only think they are.

ICARUS

[RoboJ1M]

Icarus Is Looking For You
Icarus Is Looking For You
ICARUS HAS FOUND YOU

That's not science

[Tony]

Someday, science will be just another thing your computer can do for you. If you want a new product, your computer will just plug into a cloud, design it, and then seek a manufacturing shop somewhere to make it and ship it to you.

That's not science, that's engineering. Science is the methodical discovery of the principles of reality through observation and abstraction. Engineering is a much simpler task, and one which AI may be able to handle in a few years. Science, on the other hand, will be a little more difficult, as it requires abstraction and creativity.

Has to be said, sorry!

[RivenAleem]

I for one welcome our new computer overlords!! Seriously though, there's no reason why an AI won't be benevolent instead of destructive. Look at Jane from the Ender series, or Data from startrek, there's just as much scope for them being a force of good as of evil.

Borg

[TranscenDev]

"We are the Borg. Lower your shields and surrender your ship." ~Ami Chicago Web Design

Bulshit bullshit bullshit

[Anonymous Coward]

Bullshit bullshit bullshit. This is a group of people looking for funding / losing perspective of their relative accomplishments / trying to predict the future by extrapolating from a highly edited version of the past.

First thing AI needs to do to be taken seriously is develop a machine that can read a newspaper and understand it and reason about the stories it reads at the same level a college freshman can. They are presently so far away from that it's pathetic, and it's not like they haven't been trying for the past 50 years. They can't get past basic mistakes in understanding the simplest things that any 5 year old understands about the world and never mind drawing out implications of anything. Believe me, the CIA has been wanting this for decades now.

The problem is, to interact in the world a human being would, you have to understand it as a human being would. That means you have to BE a human being- that is, the particular end product of a long chain of evolution which left us with a certain body and brain interacting with a certain environment which imposes certain constraints on that body and brain which results in certain proclivities, desires, perspectives, values, fears, along with a huge number of other epiphenomena not directly related to survival (you like music... you like like symmetric objects) all of which exists and makes sense only because of a huge vast causal chain that starts in the idiosyncratic laws of the physics of our universe which could have been different and which in turn force a certain set of equally idiosyncratic chemical properties which could have been different, which in turn forces a certain set of biological phenomena which could have been different (say, water evaporates THIS fast in this climate) which, combined with vats and vats of pure chance in turn forced a certain unrepeatable evolutionary process which resulted in Homo sapiens and their peculiar bodies, brains and needs.

So if a computer wants to reason about something undramatic and basic- human thirst- reason, say, that Joe Blow is going to be thirsty after 12 hours of no water, you can either hard code facts about people and thirst into your AI (as Doug Lenat does) and hope that those facts are sufficient to derive the proposition that Joe is likely thirsty under all appropriate circumstances (which it will not BTW because thirst is related to the amount of time spent without drinking and that is not stated but only implied by the activity .... it takes longer to get your license at the Department of Motor Vehicles than to tie your shoe) or you can try to create a program that has the same thirst needs as humans and "knows" when people are thirsty for the same reason people know they're thirsty.

The problem with the later is the exploding number of things you have to model to get it right. You have to create a model of human thirst that takes into account anything that can impact human thirst- weather conditions, salinity of the water, activity level, relative humidity.. you have to model ALL of those things and whatever else they depend on in turn. Either that, or you have only a rough approximation of human thirst. Obviously, you end up with a rough approximation.

Now start multiplying the errors of all your equally rough approximations for every little bit of human bodily, genetic, behavioral, psychological , emotional, societal systems ... all the ten trillion very complicated systems and forces that impinge on each other and which determine what a human does, figure in to each calculation you make your error and you'll see you have not a thing worthy of being called a model or a theory or AI or anything else.

Here's the upshot of all this- very very complex systems which are the determined by inscrutable and non-repeatable historical forces are their own simplest model.

Done and done.

These people are certified idiots. The "danger" is time and energy is going to be wasted yet again by society on the latest crop of researchers who are claiming, yet again, that AI is just around the corner.. at some future point in the nearest point in the future still future enough for everyone making these claims to conveniently (and thankfully) to be dead.

Re:Revoke their degrees - Robots Rules of Order

[Anonymous Coward]

Not only is Artificial Intelligence required, but so is Artificial Volition (AV). And to balance the Artificial Volition will have to be an Artificial Morality (AM); a highway code if you will.

In fact we can use a car analogy on this one: actually, a car driver analogy. AI, specifically vision processing is up to the task of parsing various inputs into a high level description such as variable sized boxes - moving and fixed objects in the field of awareness; with associated properties of establish movement vector (for basic collision avoidance), and 'extra' info such as weight (massiveness), priority to avoid collision (e.g. soft meat body versus hard metal body) and so on. AV will be a component of a Multi Agent System that feeds the 'executive' with what to do re controlling the trajectory of the vehicle: along planned, most convenient, or emergency routes (planned route is the whole trip, convenient is where traffic jams can be avoided by turning right, and emergency routes are, for example, moving in the lane of travel to avoid being side-swiped by an erraticly operated vehicle.

Artificial Morality will have to balance the volitional induction to speed (we're behind schedule) with the obvious public safety requirement that traffic vehicles operate well within the safety envelope of the least capable but street legal vehicle. So the question is, does the use of 'Robots Rules of Order' result in adherence to the legal speed limit, or perhaps the 'speed-limit upto ten over' that is common where I live, or perhaps the stealthy/non-stealthy use of speed (radar) detection equipment. Nevermind that traffic cams with ALPR (automatic licence plate recognition) can supplant the use of radar-based speed control. Travel times could be forced to be within limits for (most) vehicles.

Interesting to note that for the autonomous vehicle design competition Grand Challenge for 2008 'The Urban Challenge', detection and avoidance of pedestrians was 'not in scope.'

Mod parent up.

[FiloEleven]

Thank you for writing such an excellent post. Abstraction is very useful, but when too heavily relied upon it results in unclear thinking with similarly clouded results.

Comment removed

[account_deleted]

Comment removed based on user account deletion