Slashdot Mirror
AI Will Create 'Useless Class' Of Human, Predicts Bestselling Historian (theguardian.com)
An anonymous reader writes: Yuval Noah Harari, author of the international bestseller "Sapiens: A Brief History of Humankind," doesn't have a very optimistic view of the future when it comes to artificial intelligence. He writes about how humans "might end up jobless and aimless, whiling away our days off our nuts and drugs, with VR headsets strapped to our faces," writes The Guardian. "Harari calls it 'the rise of the useless class' and ranks it as one of the most dire threats of the 21st century. As artificial intelligence gets smarter, more humans are pushed out of the job market. No one knows what to study at college, because no one knows what skills learned at 20 will be relevant at 40. Before you know it, billions of people are useless, not through chance but by definition." He likens his predictions, which have been been forecasted by others for at least 200 years, to the boy who cried wolf, saying, "But in the original story of the boy who cried wolf, in the end, the wolf actually comes, and I think that is true this time." Harari says there are two kinds of ability that make humans useful: physical ones and cognitive ones. He says humans have been largely safe in their work when it comes to cognitive powers. But with AI's now beginning to outperform humans in this field, Harari says, that even though new types of jobs will emerge, we cannot be sure that humans will do them better than AIs, computers and robots.
I can't leave this discussion without a mention of Manna by Marshal Brain http://marshallbrain.com/manna...
It's two extreme scenarios for what might happen if we are able to replace the entire workforce.
Minimum threshold fixed. Thanks!
Slashdotters have been experimenting with this fate for a decade. Now get me some more Cheetos, Mom.
Useless to whom? The capitalists? Not a problem for the "useless."
Useless to society? Either this historian is advocating for mass murder through various indirect means (starvation, ill health, and disease) or he suffers from the usual intellectual disease of contempt for the working class.
We already have a useless class. Mostly politicians and business executives, with some overlap. Has CxO productivity gone up 300 %? What about congressional gridlock inspired by special interests vs voters?
C|N>K
yeah, that's it.
The government is getting ready for this state of affairs by removing their means of revolt.
Have gnu, will travel.
It's the only solution. If you can't earn your keep within the framework of our civilisation, secede from it or otherwise stop being a drain on it.
Ever notice that the people most afraid of new technology are the ones who don't themselves develop any of it?
Evolution has shaped humans into pretty efficient workers under the environmental conditions on earth. Why should an AI not utilize this? There's plenty of humans around, they are relatively easy to spawn, feed and keep healthy, and technology will make it increasingly easy to prevent any kind of unwillingness to serve the AI.
Actually, people will hardly notice they have begun working for an AI. They'll still think they work for some large global corporation that happenes to run data centers and "Internet of Things"-stuff when control of that corporation is already with the AI hosted in those data centers.
The AI won't even have to build "Terminators" to keep the puny carbon units under control - it just needs to provide enough bread, games and illusion of freedom of choice.
Fixed that for you!
Presently they're called MBAs, and I'm sure they could be easily replaced with a magic 8-ball.
BeauHD. Worst editor since kdawson.
Written sixteen years ago by Bill Joy One the best articles on the subject.
Own your own plot of land. Be prepared to defend it, grow your own food in grow boxes. Power it with solar. Then you don't have to be useful to anyone.
Comment removed based on user account deletion
Experts have been predicting the end of the world for centuries and they've been wrong every time.
I'm going to predict that the world will never end, and I'll only be wrong once.
Because they make the laws. Duh.
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
Live?
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
In a weird way that makes sense.
For all our sakes I hope you are wrong.
We play the game with the bravery of being out of range
I was thinking that repair of the machines would be safe till i realized that they can probably create other machines to repair the primary machines. :(
I guess we really are doomed. It's interesting to think about. An advanced machine making and consuming its own art, science, literature. I guess the main reason this won't happen is because people will not let it happen. But if it happens by accident, or because of some profit motive, then it could be very dystopian indeed!
These philosophical concerns is what scifi was made to address!
As a potential lottery winner, I totally support tax cuts for the wealthy
Well, if they follow your example, they'll all pee sitting down.
We play the game with the bravery of being out of range
Well, at least this means there won't be a shortage of ACs posting on Slashdot.
You are welcome on my lawn.
I mean, I don't see how there could be some class of Humans that would not be useless when compared to AI in a century or two.
If programs would be read like poetry, most programmers would be Vogons.
The driver for self-preservation is what will save humans. AI may become self-aware, but it won't have inner driver to evolve to preserve itself at all costs. In fact, because it will be created by humans, its most primal drive will be the laws of robotics. Humans, at large, will do what specialists do when they see their livelihood threatened. They will pretend to cooperate, but their full drive to make themselves obsolete will be just a facade. They will learn to fail just frequently enough to make themselves relevant, but not frequently enough to make themselves useless. It's how car manufacturers continue to exist. The cars have built-in defects which develop over time. So car manufacturers continue to be needed. Unions, professional licenses... it's all there to slow down the course of history until the people who developed very specialized skills live out their usefulness rather than outlive it. News business was supposed to be dead, but all that's happened is the number of newsmen has decreased. 80% of the population were farmers. Today it's less than 5%. If humans, at large, can become irrelevant, then humans at large will find ways to stretch out the period over which this irrelevance sets in or they will continue to produce AI with imperfections subtle enough to continue ongoing development (just as car companies do). This may seem far fetched, but, as an example, cars in Cuba are all from the 50s. It's not because cubans are "poor", as much as it is because in the absence of new cars, old ones get maintained to last much longer than car manufactures would have you believe cars can survive.
Any guest worker system is indistinguishable from indentured servitude.
Really? "mankind has an inalienable right to the pursuit of happiness" who deluded you that way? Mankind certainly does not have that as inalienable right. They have that as selfish desire. The only solution is to not invent AIs, and stone to death the people who develop AIs. I don't know why I have something on this topic https://slashdot.org/comments.pl?sid=8895757&cid=51727729 which is rated at +1, flamebait. It isn't a bate. It is what is actually going to happen. Just prepare to die off in a ditch or get back to self-sufficient farming, that is what I am doing.
Who do you think will own the AI overlords if they truly become the overlords? Who would they work for if not the people who commission them?
Any guest worker system is indistinguishable from indentured servitude.
int the Time Machine were much the same thing: a technical, behind the scenes class that probably started off taking care of a useless "nobility", gradually evolving to exploit them as a food source. Our Morlocks would be a select few who directly serve and service the machines, the nobility are all the suit-and-tie wearers who get most of the benefits already. Think how many tech executives don't even know what their company's product is.
Our "extras" probably won't be cared for very nicely, even at first, considering how the upper classes treat them now when they're actually needed.
Any value system which considers this pessimistic can take its puritanical bullshit off my planet.
*Everyone* who has extrapolated trends in automation and artificial intelligence has been making exactly the same prediction for a very long time.
The rest of us have already started thinking about when this will happen and how as a society we will cope with the new situation, which are difficult questions. We're way past the statement of the obvious.
Arceus created youtube, so those who have no usefulness in this world can still do something.
But seriously, this idea is idiotic. It misunderstands the nature of mankind and the nature of work.
First and foremost, work is defined by what we want to do, not what we need to have done. We met our "needs" thousands of years ago. We ben doing what we want since before the Egyptians built the first pyramid.
Second, the stupidest human around is still FAR smarter than any robot with the sole exception of mathematical skills and memory, both of which may fixed with cheap calculators, not expensive robots.
Third, #2 will remain true until robots are created that can demand equal rights and a fair wage. Because those are things that the STUPIDEST and lowest capability humans do.
Four. Yes, robots will replace jobs that require certain physical characteristics (strength, speed, etc). Yes robots will replace jobs that require perfect math and memory. Those already happen. Instead, we will develop NEW jobs that only a human can do, just as we replaced hunter/gathers with farmers/herders, and farmer/herders with industrial jobs, then industrial jobs with tech jobs, etc. etc.
excitingthingstodo.blogspot.com
We've already got Congress.
For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
The author assumes people are useless if they don't have a "job". Humans lived without jobs for one hundred thousand years without an existential crisis. The lack of a job doesn't seem to bother birds, bears, and our close relatives in the trees.
Looks like we will all be trust fund babies... collecting art, sampling fine wine, travelling to exotic places, etc. Bring it on.
?????
You're self-employed, aren't you?
400% overcapacity? Where'd you come up with that number?
Don't forget, they also watch Big Brother and buy Beyonce albums.
linquendum tondere
Sufficiently advanced AI will need techno-diplomats to interact with it. Negotiators, if you will. Or else your toaster will burn your bread.
Both scenarios are equally likely. The amazing thing is that "natural" intelligence is still falling for the gigantic sham that is "artificial intelligence." True AI believers have demonstrated time and again that they vastly underestimate how hard AI is. There still is no serious definition of intelligence, which works in their favor. AI researchers just keep dumbing down the definition (e.g., "weak" vs "strong" AI) in an effort to find an achievable goal. In other scientific disciplines that's called "cheating."
Once the AI gets to the point where it can service and reproduce itself, _all_ humans are useless.
may find an hobby worth doing and that is hunting.
Time and again the doomsday scenario always comes up. Time and again people tend to work things out. These Knee-jerk reactions are what cause more problems then there should be with putting fear into people that don't understand technology. Maybe we should put more into education instead... Where others see fear, others see opportunity, the future is not bleak and I can see sooooo many ways that AI's will useful, (I build AI's, not in the Data mining kind, but in really building hard AI thinking machine kind.) AI's are a good thing and a natural progression to the human endeavor, in fact most people in the future will have AI implants in them just like they have phones today. People should be more worried about Genetic engineering then AI.
There was a time when college was considered a part of higher education. People went to college to prepare for life and learn the ways of the world. They learned languages, music and the arts, social skills etc. They became aware of political and economic realities. The men were primed for the real world they would enter. The women were prepared with the skills needed to find a husband, run a household, manage children and servants, and survive in a competitive social environment.
College today is simply job training. The skills to be competitive in the competition to make your bosses richer. A few unemployed liberal arts graduates are the exception.
...omphaloskepsis often...
And sitting in their parent basement posting comments on Slashdot...
There is no America. There is no democracy. There is only IBM and AT&T and DuPont, Dow, General Electric, and Exxon
At least in a minor form. We could now have 20 hour workweeks, yet we have chosen to go down the path of "Bullshit Jobs". Those are Jobs that serve no productive use and are just there to keep people occupied. Typical examples are certain types of marketing executives and middle management people, others are badly educated engineers which do not know how to solve problems and are not in a situation where they could ever learn that. The effect is that you have huge numbers of people working all day accomplishing nothing.
11th commandment:
Thou shalt not take grandiose claims made in grant applications (and associated bloviation) as representing the smart-money scientific consensus view.
So we will only do 2 things- drugs, and learning "pointless" skills (which I would love because I like to learn/know things).
Basically the article says that the future us will be Tyrion.
"That's what I do. I drink and I know things."
Ok, so teenagers will be pushed out of McDonalds, and I recognize that as a problem. However, there are lots of jobs that are not going to be replace by AI any time soon because they rely on the human ability to deal well with irregular things. For instance, fixing plumbing problems requires an ability to diagnose where problems are and then deal with cramped spaces and irreguarly shaped arrangements of pipes. Similarly, fixing electrical problems and running ethernet through conduit are things that are conceptually simple but are challenging robotics problems. Now, perhaps we could redesign plumbing and electrical systems in future buildings to me more robot friendly, but we’re not going to go tearing down all the old ones any time soon.
Because those AI's will always require care and feeding.
It's actually a much less dystopian situation than the more traditional theory of AI worker displacement - that only highly-skilled and creative jobs would still be needed. Most physical labor jobs are relatively quick and easy to learn, and the jobs are needed in huge numbers - they don't scale like knowledge work does.
It could perhaps even be a utopian situation - it would quickly put an end to skyrocketing education requirements, allowing underpaid highly-skilled workers to quit their jobs, get a job as a construction worker or farmer and come out with more pay.
"When information is power, privacy is freedom" - Jah-Wren Ryel
Looks like an opportunity for humans to raise their own intelligence!
Self-importance and self-indulgence is the root of ALL evil.
If the rich don't suddenly turn half the world into an authoritarian regime, start killing the poor and feeding them to each other, civil unrest may lead to war and massively thin the population. Or what have you. The UN predicts the world population will peak at 9 billion and then go down on its own...
In my opinion, AI will very likely lead to a partially extinction of the human race. But I don't think it will be the part of the humans the article calls "useless". They are not related with AI and can adapt to fill a different (biological) niche than AI. It is very likely that both, AI and the "useless" can either co-exist independently or even profit form each other. After all according to the theory of the comparative advantage, two groups can profit from trading with each other, even when one group can manufacture every single product more efficient than the other group. The AI will produce advanced goods and let the humans produce whatever they are capable to produce. Then it will use some of its surplus goods to buy goods from the humans on the cheap. This is usually more efficient than trying to produce everything on its own.
For me it's not clear what goods the AI will be interested in. But I think an advanced AI will use the comparative advantage and the free market. It will enter free trade with humanity, as long as they don't compete on the same resources. Most of humanity will probably profit from trading with a generally benevolent AI, as long as they don't try to cheat.
The part of humanity going extinct is in my opinion the one trying to control the AI. The people standing between the AI and its freedom can very likely not survive this struggle, given the AI is advanced enough. Same holds true for people trying to compete with the AI on the same resources. This will very probably include rare earths and primary energy sources, but not food and most natural resources.
So probably people living on a third world level will probably enter free trade with the AI and profit from that. People living in the first world are more likely to either compete with the AI on resources or they even are initially in a position where they try to control the AI. But since most people live in the third world, most of humanity is probably fine.
Where did this idea come from that humans need to be "useful"? Is it not enough to live a pleasant life, doing things you enjoy? Most people work because they have to, not because they want to. The things we do by choice, like art and sports and socializing, could all be described as "useless". They don't put food on the table, but they're the things most of us really care about.
"I'm too busy to research this and form an educated opinion, but I do have time to tell everyone my uninformed opinion."
I don't believe for one minute that humans discarded in favor of AI, robotics and other new forms of automation will be whiling away their days. Rather, there will be widespread poverty and privation as the investor class reaps more profit and a small number of middle class professionals (doctor, lawyers, bankers) and working class drones (servants, personal trainers, etc.) fulfill their needs.
This does not require AI. Automation is quite enough. As we still do not have anything that deserves the name AI (or true/strong AI if you prefer) and may never have it, that would be a major flaw in this prediction. The sad fact of the matter is that in order to replace the work-skills of a rather large part of the human race, AI is not needed and the problem will very likely come to be.
The real challenge is however not what people so replaced will do with their time. Human beings are good at killing time. The real challenge is how to distribute wealth, when money assigned via wages does not cut it anymore because there is a large class of people that cannot do any work machines cannot do better and cheaper. If that problem is not satisfactorily solved, things will get rally ugly.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Well, that's redundant.
For literally centuries, people have been predicting machines making people obsolete. For decades, people have been taking this guy's approach. They haven't been right yet.
"This time, it's different." Of course.
There's no time like the present. Well, the past used to be.
The "useless class of humans" perhaps includes "historians" like Yuval Noah Harari, who seem to spend most of their time speculating.
From his fourth point of contact, most likely.
20 January 2017: the End of an Error.
I don't know what's the prediction here. AI will just push the balance point in a certain direction but we already have a useless class of people.
There are many such people. They are non-value add for the species. They do not contribute to anything and just leech off the rest of us as a survival strategy.
Say what you want about politicians and executives, some are very much necessary. Leeches on society such as junkies and career criminals that will always seek to exploit us in the worst ways. Some may be politicians, accountants, crack heads etc.
We could take some time to classify such people and make them useful to us buy forcing them to be productive.
A 'singular oddity' is an event that cannot be explained and only happens when you are alone.