Most people not deeply involved in the artificial intelligence frenzy may not have noticed, but perceptions of AI\u2019s relentless march toward becoming more intelligent than humans, even becoming a threat to humanity, came to a screeching halt Aug. 7.<\/p>\n
That was the day when the most widely followed AI company, OpenAI, released GPT-5, an advanced product that the firm had long promised would put competitors to shame and launch a new revolution in this purportedly revolutionary technology.<\/p>\n
As it happened, GPT-5 was a bust. It turned out to be less user-friendly and in many ways less capable than its predecessors in OpenAI\u2019s arsenal. It made the same sort of risible errors in answering users\u2019 prompts, was no better in math (or even worse), and not at all the advance that OpenAI and its chief executive, Sam Altman, had been talking up.<\/p>\n
AI companies are really buoying the American economy right now, and it\u2019s looking very bubble-shaped.<\/p>\n
\u2014 Alex Hanna, co-author, \u201cThe AI Con\u201d<\/p>\n
\u201cThe thought was that this growth would be exponential,\u201d says Alex Hanna, a technology critic and co-author (with Emily M. Bender of the University of Washington) of the indispensable new book \u201cThe AI Con<\/a>: How to Fight Big Tech\u2019s Hype and Create the Future We Want.\u201d Instead, Hanna says, \u201cWe\u2019re hitting a wall.\u201d <\/p>\n The consequences go beyond how so many business leaders and ordinary Americans have been led to expect, even fear, the penetration of AI into our lives. Hundreds of billions of dollars have been invested by venture capitalists and major corporations such as Google, Amazon and Microsoft in OpenAI and its multitude of fellow AI labs, even though none of the AI labs has turned a profit. <\/p>\n Newsletter <\/p>\n Get the latest from Michael Hiltzik<\/p>\n Commentary on economics and more from a Pulitzer Prize winner.<\/p>\n Enter email address <\/p>\n Sign Me Up <\/p>\n You may occasionally receive promotional content from the Los Angeles Times. <\/p>\n Public companies have scurried to announce AI investments or claim AI capabilities for their products in the hope of turbocharging their share prices, much as an earlier generation of businesses promoted themselves as \u201cdot-coms\u201d in the 1990s to look more glittery in investors\u2019 eyes. <\/p>\n Nvidia, the maker of a high-powered chip powering AI research, plays almost the same role as a stock market leader that Intel Corp., another chip-maker, played in the 1990s \u2014 helping to prop up the bull market in equities.<\/p>\n If the promise of AI turns out to be as much of a mirage as dot-coms did, stock investors may face a painful reckoning.<\/p>\n The cheerless rollout of GPT-5 could bring the day of reckoning closer. \u201cAI companies are really buoying the American economy right now, and it\u2019s looking very bubble-shaped,\u201d Hanna told me. <\/p>\n The rollout was so disappointing that it shined a spotlight on the degree that the whole AI industry has been dependent on hype. <\/p>\n Here\u2019s Altman, speaking just before the unveiling of GPT-5, comparing it with its immediate predecessor, GPT-4o: \u201cGPT-4o maybe it was like talking to a college student,\u201d he said. \u201cWith GPT-5 now it\u2019s like talking to an expert \u2014 a legitimate PhD-level expert in anything any area you need on demand … whatever your goals are.\u201d <\/p>\n Well, not so much. When one user asked it to produce a map of the U.S. with all the states labeled, GPT-5 extruded a fantasyland<\/a>, including states such as Tonnessee, Mississipo and West Wigina. Another prompted the model for a list of the first 12 presidents, with names and pictures. It only came up with nine<\/a>, including presidents Gearge Washington, John Quincy Adama and Thomason Jefferson. <\/p>\n Experienced users of the new version\u2019s predecessor models were appalled, not least by OpenAI\u2019s decision to shut down access to its older versions and force users to rely on the new one. \u201cGPT5 is horrible<\/a>,\u201d wrote a user on Reddit. \u201cShort replies that are insufficient, more obnoxious ai stylized talking, less \u2018personality\u2019 \u2026 and we don\u2019t have the option to just use other models.\u201d (OpenAI quickly relented, reopening access to the older versions.)<\/p>\n The tech media was also unimpressed. \u201cA bit of a dud<\/a>,\u201d judged the website Futurism and Ars Technica termed the rollout \u201ca big mess.\u201d<\/a> I asked OpenAI to comment on the dismal public reaction to GPT-5, but didn\u2019t hear back.<\/p>\n None of this means that the hype machine underpinning most public expectations of AI has taken a breather. Rather, it remains in overdrive. <\/p>\n A projection of AI\u2019s development over the coming years published by something called the AI Futures Project under the title \u201cAI 2027\u201d<\/a> states: \u201cWe predict that the impact of superhuman AI over the next decade will be enormous, exceeding that of the Industrial Revolution.\u201d <\/p>\n The rest of the document, mapping a course to late 2027 when an AI agent \u201cfinally understands its own cognition,\u201d is so loopily over the top that I wondered whether it wasn\u2019t meant as a parody of excessive AI hype. I asked its creators if that was so, but haven\u2019t received a reply.<\/p>\n One problem underscored by GPT-5\u2019s underwhelming rollout is that it exploded one of the most cherished principles of the AI world, which is that \u201cscaling up\u201d \u2014 endowing the technology with more computing power and more data \u2014 would bring the grail of artificial general intelligence, or AGI, ever closer to reality. <\/p>\n That\u2019s the principle undergirding the AI industry\u2019s vast expenditures on data centers and high-performance chips. The demand for more data and more data-crunching capabilities will require about $3 trillion in capital<\/a> just by 2028, in the estimation of Morgan Stanley. That would outstrip the capacity of the global credit and derivative securities markets. But if AI won\u2019t scale up, most if not all that money will be wasted.<\/p>\n As Bender and Hanna point out in their book, AI promoters have kept investors and followers enthralled by relying on a vague public understanding of the term \u201cintelligence.\u201d AI bots seem intelligent, because they\u2019ve achieved the ability to seem coherent in their use of language. But that\u2019s different from cognition. <\/p>\n \u201cSo we\u2019re imagining a mind behind the words,\u201d Hanna says, \u201cand that becomes associated with consciousness or intelligence. But the notion of general intelligence is not really well-defined.\u201d <\/p>\n Indeed, as long ago as the 1960s, that phenomenon was noticed by Joseph Weizenbaum, the designer of the pioneering chatbot ELIZA, which replicated the responses of a psychotherapist so convincingly that even test subjects who knew they were conversing with a machine thought it displayed emotions and empathy.<\/p>\n