Daron Acemoglu , wants to make evident right away that he has little against artificial intelligence. He gets the ability. ” I’m not an AI pessimist”, he declares moments into an exam.
The unending hype surrounding the technology and the method it’s fueling an investment boom and angry tech stock rally makes Acemoglu, a distinguished professor at , Massachusetts Institute of Technology, come off as a doomsayer locked in on the mounting economic and financial perils away.
As promising as AI does remain, there’s little chance it will live up to that enthusiasm, Acemoglu says. By his analysis, just a small percentage of all tasks — a mere 5 % — is ripe to be taken over, or at least seriously aided, by AI over the next generation. True, good news for workers, but very poor for the businesses that are pouring billions into engineering in hopes of seeing a rise in productivity.
” A lot of money is going to find wasted”, says Acemoglu. You wo n’t experience an economic revolution with those 5 %, they say.
Acemoglu has become one of the louder, and more high-profile, tones warning that the AI panic on , Wall Street , and in C-suites across America has gone too far. Acemoglu, a professor at the Institute of Technology and the highest title for faculty at MIT, made his name a decade ago when he co-authored the New York Times bestseller” Why Nations Fail.” AI, and the advent of new technologies, more widely, have figured strongly in his finance work for years.
The opponents contend that AI will enable companies to manage a significant portion of daily tasks and set off a new century of medical and scientific advancements as the technology continues to develop. According to Jensen Huang, CEO of Nvidia, a company whose very name has become synonymous with the AI boom, rising demand for the technology’s services from a wider range of businesses and governments will require as much as$ 1 trillion in spending to upgrade data center equipment in the upcoming years.
Suspicion about these kinds of claims has risen, in part because assets in AI have increased costs significantly faster than earnings at companies like Microsoft and Amazon, but most investors still want to pay astronomical premiums for shares poised to drive the Artificial wave.
In the upcoming years, Acemoglu envisions three possible approaches the Artificial history might unfold.
The second, and by far the most favorable, situation calls for gradual cooling of the hype and widespread funding for “modest” uses of technology.
— In the second scenario, the fury builds for another year or therefore, leading to a tech stock accident that leaves owners, professionals and students disillusioned with the systems. ” Artificial spring followed by AI wintertime”, he calls this one.
The fourth and worst scenario is that the mania continues unregulated for decades, leading to the elimination of thousands of jobs and the pouring of hundreds of billions of dollars into AI “without understanding what they’re going to do with it,” leaving employers scrambling to rehire employees when the technology fails. ” Now there are common bad outcomes for the entire market.”
The most good? He concludes that the third and second situations are a mix. Inside C-suites, there’s just too much fear of missing out on the AI growth to foresee the publicity system slowing down any time soon, he says, and” when the buzz gets intensified, the slide is unlikely to get soft”.
Second-quarter figures illustrate the magnitude of the spending frenzy. Four companies alone — Microsoft, Alphabet, Amazon and , Meta Platforms , — invested more than$ 50 billion into capital spending in the quarter, with much of that going toward AI.
Today’s large language models like OpenAI’s ChatGPT are impressive in many respects, Acemoglu says. So why ca n’t they replace humans, or at least help them a lot, at many jobs? He points to reliability issues and a lacked human-level wisdom or judgment, which will make it unlikely that people will soon outsource many white-collar jobs to AI. He claims that AI wo n’t be able to automate physical tasks like cleaning or construction.
” You need highly reliable information or the ability of these models to faithfully implement certain steps that previous workers were doing,” he said. ” They can do that in a few places with some human supervisory oversight” — like coding—” but in most places they cannot”.
” That’s a reality check for where we are right now”, he said.
___
© 2024 Bloomberg L. P
Distributed by Tribune Content Agency, LLC.