I've long argued that eventually AI will get better than human beings at everything, which in turn means we'll all be out of jobs.¹ During the Industrial Revolution this didn't happen because while machines took over a lot of jobs, they also created a lot of new jobs for humans (like designing and maintaining machines). AI is different. If it can do anything, then by definition any new job you can think of can also be done better and cheaper by AI. It's game over.
But along comes Noah Smith with a clever counterargument. He doesn't deny that AI will improve, or even that it will eventually get better at everything. His case is more subtle.
In a nutshell, he suggests that no matter how good AI gets, it will always be valuable enough to be allocated to the highest value tasks. This means there might still be lots of jobs left for humans. Even if we're comparatively lousy at them, it could make sense to keep at them if it frees up AI for more important work.
Here's an example to make this concrete. Suppose we're invaded by aliens who are intent on killing us all. This is an existential threat, and it would therefore make sense to dedicate 100% of our compute power to fighting the aliens. A robot might still be a better farmer than a human, but we'd all grow our own food if it meant increasing the number of robots defending the planet.
So how likely is this? The most obvious real-world answer is that compute power is likely to grow so much that every human task can be done by a fraction of a percent of the world's total AI—and the more AI grows, the tinier the human fraction gets. Technically this doesn't matter: in the case of the alien invasion, for example, you'd still want to use every last petaflop of compute on fighting back no matter how much you had.
But that sort of existential threat is fanciful. In the real world, there are always lots of frictions and adjustments. It seems unlikely that we'd all keep working just because, technically, that last 0.01% of compute power could be put to better use. It would have to be a helluva better use, no? An improvement of 1% in GDP wouldn't cut it.
So it's a nice argument, but I don't buy it. It seems vanishingly unlikely that, politically, we'd condemn ourselves to lives of drudgery based on an ultra-purist free-market promise that it's for the best. We certainly never have before.
¹Or 99% of everything if you insist on believing that chemical computers will always be able to do a few things better than digital computers. It doesn't matter. It's mass unemployment either way.
where AI is best suited and would save the most money is CFO and CEO plus upper management. I foresee most of AI computer cycles bought by CFOs and CEOs having AI present arguments whey they should keep their jobs instead of AI.
I provisionally agree with you. These upper echelon types are (IMHO) essentially -- supposedly -- functioning as diagnosticians, albeit business diagnosticians as opposed to medical diagnosticians, employing the same feature recognition techniques as those used in say AI Radiology to diagnose lung cancer, brain tumors, etc.
IOW, they can be replaced and easily at that by an algorithm once you've tuned all the (hyper) parameters / features, etc.
AI, at least now, is not analytical. It does not think and plan. It is a glorified Markov chain based on what has already been written. As such, it is not capable of planning for a future that is not already anticipated and discussed in the training data it is given.
Given this, an AI is not capable of the strategic planning that is a CEO's most important task. It would not have made Apple design the iPod or made Coca Cola introduce New Coke (which contrary to popular belief was a huge success for Coca Cola, cementing the image of the original Coca Cola as America's drink).
Absolutely! Upper management jobs are the most amenable to automation, since, at best, they are involved in analyzing easily quantifiable data. I thought so even before it was "formally" announced that there was AI.
Another profession that stands to improve with AI is medicine.
Help me out here. If we do get the point that AI can do pretty much everything, what reason would the rich and powerful have for keeping the proleteriat around? If we are no longer necessary to provide them with the stuff they accumumulate, what then is our utility? And why would they give us a universal income?
They would have no reason. The "proletariat" would be left to fend for themselves, without jobs or income of any kind (once all the so-called safety nets expire, which would be a couple of years), and without survival skills (a highly-trained doctor, lawyer or programmer would not know how to grow food). The .1% wouldn't waste any energy exterminating the rabble. They would just close the gates and those of us outside would kill each other and then die off.
No, the question is why would the proletariat allow "rich and powerful" to continue to be a meaningful concept?
I think the proletariat will be fine. It's upper and middle management, doctors, lawyers, that are in danger. Unless AI can pump out a septic tank or redo my living room, the proletariat is still needed.
Pingback: Will AI take all the jobs? - Angry Bear
AI and robots are the same thing now? And KD has once again not gotten the memo that Moore's Law has been repealed. Computer power is not growing like it once did. Not even close.
We are hitting the physical limits of how small we can make circuits. But we have gotten better at making larger silicon wafers, larger chips, stacking components, and linking chips together.
Moore's law has "evolved". The old days of shrinking dyes, packing more transistors into the same area, and getting much more bang for the buck (cheaper to make, cheaper to run and more powerful!) are over.
Yes, computational power is growing horizontally, not vertically. Advancements in AI are due a much greater increase in memory, storage capacity, and speed of retrieval.
Someone once asked in disbelief why the Star War universe had human-like robots, but also slaves to do manual labor.
The answer is to simply look at our world. At this moment there are robotic surgeons doing astonishingly complex high value tasks, and also 13 year old children doing manual labor like picking lettuce.
Because machines are more expensive than humans.
I've said for a long time that AI will be to white collar labor in the 21st century what electro-motive power was to blue collar labor in the 20th.
More and more, the white collar professions are seeing the high value tasks taken by machine algorithm, while the human becomes the machine tender.
It's not really a question of what tasks AI will do, but how much will be paid for them. Not all that long ago people paid people to do their laundry. Then the washing machine came, laundry became CHEAP, and the money was spent on other things humans produce. Right now ChatGPT can undoubtedly produce a better speech than Hillary Clinton was ever paid 100,000 $ for. For free. So the hundred grand will be spent on something else, won't it? (Except if it's STILL spent on Hillary Clinton, but for her presence not her speech.)
I worry AI will exacerbate income inequality, hugely. That it will also lead to a race to the bottom as capital, freed from labor, will go wherever taxes are lowest. But will AI take away our jobs? Sure, but then those jobs won't be worth much anymore and others will. The trick will be to keep enough money in ordinary human hands to let it be spent on other human hands.
Trying to imagine a world in which AI has taken all but one percent of the available paying jobs...
Unemployment benefits -- in those countries that have them -- would be depleted in a few years tops. Starting on the day Kevin thinks AI will begin taking all our jobs, in ten years at least a quarter of workers in the developed world will be scrambling for new, lesser, jobs. (That's already the case in the third world, not because of AI.) Ten years after that, it will be all of us.
The super-rich won't care, of course. Their need for doctors, lawyers, technicians will employ a only minuscule number of those kinds of skilled jobs, and competition for them will be cutthroat, thus they will be paid and treated like servants. AI will help to develop really hot artificial sex partners, willing to perform literally any act imaginable (but only available to billionaires), so even the Oldest Profession will come to an end.
Outside the mansions and gated communities of the .1%, life will be brutal and short. With rampant looting, robbery and other property crimes, robot law enforcement would put most of the rabble in prison, but only if the rich folks see the problem coming, which they won't. They also won't notice that their doctors and other skilled slaves will eventually die and not be replaced, so things will get kind of crappy for the upper class, too.
Civilization won't last past the end of this century, and human life on this planet will devolve into small tribes of hunter-gatherers. Maybe there'll be a craze of rich people actually venturing out of their mansions to sport-hunt the remaining humans -- using helicopters and heat-seeking ammunition -- or maybe they'll stay in their mansions and hunt using drones.
Somewhere along the way AI will make it feasible for (rich) people to leave this planet, and some will do so. If they find other planets that can support life, they can start a new civilization built entirely on greed and self aggrandizement. Except they won't want to procreate, because the AI sex partners will have jaded them to the charms of human intercourse and they will have no interest in doing the things that cause babies. So that "civilization" will die out quickly.
I feel like the "take our jobs" thing is conflating "AI" with "robotics". Robotics have a very real cost to building and operating them, and it's not clear that they'd always be cheaper in a world where AI could do all the brain work that humans can.
Today's robots are designed and built to perform one task: attach the front bumper to the Hyundai. They won't do the back bumper, or any bumper on a Honda. But endow them with a little "intelligence" and the same robot will be able to figure out how to do multiple things. This would make them more affordable, I think, as they could be assigned wherever they were needed. So I'd expect a merger between AI and robotics.
My question is who would buy all those cars, once we're all out of work?
While increasing inequality might push all of the prols into "developing country" conditions, the companies that make the AI/robots will still be driven by greed which will keep the cost of the robots at or above the cost of human labor. No one will give away the robots. Hence there will generally be crappy menial jobs for many if not all of the human dregs.
Besides, what's the fun in being wealthy if you don't have the poor around to compare yourself to??