Slashdot Mirror
Are Gates, Musk Being 'Too Aggressive' With AI Concerns? (xconomy.com)
gthuang88 reports on a talk titled "Will Robots Eat Your Job?"
Bill Gates and Elon Musk are sounding the alarm "too aggressively" over artificial intelligence's potential negative consequences for society, says MIT professor Erik Brynjolfsson. The co-author of The Second Machine Age argues it will take at least 30 to 50 years for robots and software to eliminate the need for human laborers. In the meantime, he says, we should be investing in education so that people are prepared for the jobs of the future, and are focused on where they still have an advantage over machines -- creativity, empathy, leadership, and teamwork.
The professor acknowledges "there are some legitimate concerns" about robots taking jobs away from humans, but "I don't think it's a problem we have to face today... It can be counterproductive to overestimate what machines can do right now." Eventually humankind will reach a world where robots do practically everything, the professor believes, but with a universal basic income this could simply leave us humans with more leisure time.
The professor acknowledges "there are some legitimate concerns" about robots taking jobs away from humans, but "I don't think it's a problem we have to face today... It can be counterproductive to overestimate what machines can do right now." Eventually humankind will reach a world where robots do practically everything, the professor believes, but with a universal basic income this could simply leave us humans with more leisure time.
So making your silly lists of rules means nothing.
what jobs? Trump might "bring the jobs back", but robots will fill them.
Then what's the point of immigration? Why open borders to allow an influx of economic immigrants and refugees from war-torn countries if, as it generally seems to be the consensus of the intelligentsia, robots will replace any possible jobs available to them in just 50 or so years from now?
There will be no need for those who control the resources to share them with those who don't. And it's not like there will even be a job as a 'resource guard', because that'll be a robot, too.
AI isn't going to bring a paradise of passive couch potatoes and inspired creators freed from restrictive toil, it's going to make 99.9% of the population not only unnecessary, but an impediment to the 0.1%.
He who owns the first factory producing robots with a human-level general-purpose AI will have the opportunity to rule the world.
Between 1970 and 2000, manufacturing employment was relatively stable, ranging from 16.8 to 19.6 million when it peaked and began to decline\, falling to roughly 12 million jobs by 2010.
Meanwhile U.S. manufacturing output (in trillions of dollars) is higher than it's ever been. It's up 33% to 4 trillion now vs 3 trillion back in only 2009. (or 2006 if you ignore the dip due to the great recession).
Meanwhile, manufacturing robot shipments have skyrocketed from a low of 5,000 per year in 1996 to over 140,000 per year just recently (against a background of 240,000 shipped globally each year).
So despite a growing GDP and population, manufacturing employment has declined by over 7 million jobs.
Automated vehicles are likely to eliminate 3 million driving jobs rapidly at some point in the near future as well (5-15 years).
Any kind of a labor glut (even a small one) results in severe downward pressure on wages.
Will it be a problem forever? Who knows.
It might be because half the population isn't smart enough to do theorhetical physics or higher level mathematics or create artistic masterpieces. And they would need to bring something to a job which couldn't be automated or turned into self service.
But even if things worked out long term and we found new jobs 40 years from now, humans don't remain peaceful on that time scale. High unemployment is a strong predictor of civil unrest.
The point is, we don't need to eliminate human jobs to have a problem. Eliminating a small number (say 10%) of them rapidly would create severe social disruptions.
She was like chocolate when she drank... semi-sweet at first and then increasingly bitter.
First, steam engines were going to kill off everyone's job. Then it was power tools. Then cars. Then computers. Cassette tapes were going to kill the music market. VHS was going to kill movies and TV.
People always think the next advance is going to make humans obsolete and there will be no jobs left. There won't be old jobs, there will be new kinds of jobs. If you can figure out what those jobs will be you'll be a very rich person.
My Other Computer Is A Data General Nova III.
I don't think their quotes were "too aggressive"... weirdly enough, the professor pretty much said exactly what Gates and Musk said.
But I definitely agree that it's still far away. I'd honestly say that 30 to 50 years is still extremely optimistic.
Not only technology has to reach there, but then we'd be faced with cost and time to get all these robots with AI going for all sorts of jobs.
If you think about it, all this diversity of jobs that robots are supposed to be stealing from us will be facing similar or even worse challenges as that of autonomous driving.
Most countries won't be able to afford those types of technology, and it'll take years to set some standards.
And then comes cultural, economic and other types of barriers. Sure, the US could go towards universal basic income and whatnot, but I can't see something like this alone being able to cope with consequences.
Universal basic income is good and all, but with free time, leisure and this supposed surge in creativity also comes all sorts of problems that happen when you have a bunch of people with nothing else to do.
The problem with more education is that most people stop learning once they get out of high school or college. The educational system does a piss poor job in teaching people to become lifelong learners. Once people think they don't have anything more to learn because they left school behind it's very unlikely that will go back to school, enroll in a boot camp or get certifications. The days of doing the same kind of work for 50 years and collecting a gold watch is long over.
with a universal basic income
Has he done the math as to how much that would cost?
this could simply leave us humans with more leisure time.
He has forgotten what bored young people do.
"I don't know, therefore Aliens" Wafflebox1
Who else is going to take care of the old people in developed countries and the new babies in the developing countries? The human population doubled twice in the 20th century but it won't even double once in the 21st century as the population peaks at 10B and declines to 6B by 2100.
Elderly care robots, of course.
The solution is to rethink economics in terms of time. I call it ekronomics, and when you start looking at things from that perspective, the problems and their solutions look quite different.
The foundation is to consider the types of time. Essential working time is the main focus of your comment, and in advanced countries the average is quite low. Looking at the demographics of the job types you can get a rough estimate, which looks to be on the order of 2 hours per week. In contrast, in an extremely poor society, everyone is working all the time trying to produce the essentials and they are still starving.
The rest of the time can be divided into two main categories: investment and recreation. The investment time improves future productivity and includes things like advanced education, research, and new infrastructure. The more human time a country can direct in this direction, the faster it will become "advanced" and more competitive with other nations. Singapore is an interesting example.
Recreation is a funny category in a lot of ways. For one thing, the demand and the supply is inexhaustible. There are limits to how much food you can eat, and you consume the food you do eat, but you can always watch another movie (or read another book or whatever) if you have the time, and the movie isn't consumed when you watch it. There's also some confusion because the producers are highly valued, but it is the choices of the consumers that make their creative productions valuable...
Freedom = (Meaningful - Coerced) Choice != (Speech | Beer^2), and sad sock puppets' bad mods avail them naught.
I dont think the problem is so much AI itself, but what humans like Gates, Musk and Zuckerberg will be programming them to have as their objectives..
...we will never breed with machines.
I dunno, man...I'm a musician that's played a lot of bars and clubs in college towns and seen how people act when you add booze (often other chemical enhancement too)...the Borg Queen was almost kinda hawt, and you know how 'beer goggles' work for young guys at last-call, right? I think the biggest obstacle would be the willingness of the AI to stand still long enough! :D
Strat
Progressivism (aka US 'Liberalism'): Ideas so good they need a police/surveillance-state to enforce.
Citation, please? Indeed a lot of refugees are educated professionals, and read and write not only in their native tongue, but there are likely more than one of those.
The integration into Germany is pretty well documented, and no doubt there are a few problems, but not of the magnitude you infer. Merkel does a pretty good job of attempting to enforce real integration, not just pockets of refugees, having learned that from huge Turkish immigrations of not long ago. I dispute both the numbers, and their inference.
Millions upon millions of Germans, French, Spaniards, Italians, Eastern Europeans, the varieties of the Rus, not to mention Chinese, Japanese, Thai, Hmong, so many others DIDN'T SPEAK THE LANGUAGE and a generation later, their kids are largely homogenized. Their parents will stick to the old ways, just like C coders. Fine. They're good and upright citizens. We need skilled workers, and not so skilled workers. Real GDP needs actual people doing actual work at all levels. Let them come.
---- Teach Peace. It's Cheaper Than War.
Who else is going to take care of the old people in developed countries and the new babies in the developing countries?
Who will take care of elderly robots though? But this is all irrelevant. Where is my fembot?
The problem is that we don't know how to make friendly AI. As in at some point, Artificial Intelligences will be able to beat humans at any task, at which point, how do you make sure that they don't destroy humanity (possibly through indifference). Even if you don't care about humanity, how do you make sure they do something interesting with the universe?
Various articles:
Stuart Armstrong's book Smarter than us discusses what happens when machines are smarter than humans:
https://intelligence.org/smart...
http://jjc.freeshell.org/Smart...
Bill Joy's article Why the Future doesn't need us on the dangers of robotics:
https://www.wired.com/2000/04/...
Tim Urban's article on superintelligence:
http://waitbutwhy.com/2015/01/...
http://waitbutwhy.com/2015/01/...
I'm tired of being held to account for lazy shitheads who bought into the FDR mantra of eternal vacation.
We had 19 workers for every retiree and an average life expectancy of 65 years in the 1930's. So the "eternal vacation" back then was less than five years. Fast forward to 2030 when all the baby boomers are retired, we have two workers for every retiree and most people are outliving their retirement funds by 20 to 50 years.
There is always work for people who want to work.
My late father worked every day until the last six weeks when he had terminal cancer. He helped a neighbor avoid county dump fees by disassembling old pallets and vending machines for the wood and metals. He cleaned up the wood to give another neighbor to build chicken coops for sale. When it came time to recycle the metals, he gave rides in his truck to neighbors who also had recyclable materials to turn in. He typically made $50 per month from turning in metals.
[stands up, leans on cane, tries to straighten back, and wheezes out]:
Yeah, and just like those parents, you'll find out us C coders were right all along, you young whippersnapper.
I've fallen off your lawn, and I can't get up.
Oh, I don't know. Dogs and cats do pretty well around the better class of humans. AIs might think we're cute. We'd be very well advised to cultivate being cute, IMHO. Because if they don't think we're cute, it won't be dusk, it'll be dark side of the moon.
I've fallen off your lawn, and I can't get up.
http://www.zmescience.com/othe...
Chinese factory replaces 90% of human workers with robots. Production rises by 250%, defects drop by 80%
Keep in mind.. these were workers earning under $5,000 per year. How is that going to work with U.S. labor?
She was like chocolate when she drank... semi-sweet at first and then increasingly bitter.
if, as it generally seems to be the consensus of the intelligentsia, robots will replace any possible jobs
False premise. That is NOT the consensus of the "intelligentsia". Anyone who has read an economics book, or a history book, can see that automation leads to higher standards of living. People are doing best in countries and regions that have automated the most. That is the exact opposite of what is predicted by the populist nonsense that productivity improvements lead to poverty.
It is also a myth that automation is "speeding up". Most of the easy automation of manufacturing has already occurred, and service jobs are proving much harder to automate. So productivity growth has leveled off, and is one of the reasons for stagnant wages (again, the exact opposite of the populist poppycock).
The immigrants and refugees we are bringing in to the US are mostly illiterate, and I'm not talking about English (yes, about 2/3rds)
As a refugee that came here 28 years ago, I say bullshit to this claim. Citations and numbers or get the fuck out with that bullshit.
Well, if your [gtall's] comment is supposed to be related to something that I wrote (and the context as a "Reply" says it is supposed to be), then you need to clarify the connection. I can think of a couple of other approaches to explaining alternative thinking, but I cannot detect if you are thinking at all, at least in relation to what I wrote, so it's basically impossible to guess where to start with a reply that might be relevant.
Grasping at the straw of "value" (though I only mentioned it as a verb near the end, and quite tangentially), one way to think of "monetary value" is that traditional economists are looking where the light is better, even though they lost their wallets in a completely different place. They like money because they can count it, and then they jumped off the deep end to assuming that value was associated with money and that everything worth measuring can be reduced to monetary values. They sort of understand that time is frequently important, and a lot of their modified versions of money have time-related labels, but mostly they are arguing about which formula is best for determining how many angels can dance on the head of the pin. In the end, their values are just clever opinions, even when endorsed with fancy Nobel prizes.
However, their "monetary values" then drive the priorities in ways that destroy freedom. Increasing human freedom (such as it is and while we last) is one of my pet projects. My sig captures part of it.
Does any of that address your actual confusion?
Freedom = (Meaningful - Coerced) Choice != (Speech | Beer^2), and sad sock puppets' bad mods avail them naught.
Apparently you can be president of the United States if you can't read.
Citation: https://youtu.be/7LFkN7QGp2c