Billions of dollars were wiped off the stock market value of US-listed software and financial companies on Monday after a Substack post outlining a bear case for artificial intelligence and the American economy went viral.
Citrini Research, an equity and macro research firm in New York, described a potential scenario for the “consequences of abundant intelligence”, in which by 2028, while the “owners of compute” saw their wealth explode as labour costs vanished, real wage growth collapsed as white-collar workers lost jobs to machines and were forced into lower-paying roles.
The article conjured up a future whereby, starting in software, technological disruption would hit ecommerce, payment processors and other sectors. The fallout would spread across the private credit and insurance industries. Meanwhile, unemployment would fuel mortgage defaults in formerly affluent housing markets. The economy would take a hit.
“In every way AI was exceeding expectations, and the market was AI,” according to the bearish scenario. “The only problem … the economy was not.”
Shares in many of the companies mentioned by name in the post fell sharply on Monday, including Visa and Mastercard, which fell 4.6 per cent and 5.8 per cent, while DoorDash dropped 8.4 per cent and Blackstone shed 6.2 per cent. IBM, one of the world’s biggest computer companies, lost 13.2 per cent.
The tech-heavy Nasdaq closed down 258.80 points, or 1.1 per cent, at 22,627.27, while the Dow Jones industrial average lost 821.91 points, or 1.7 per cent, at 48,804.06 amid a broader sell-off linked to trade uncertainty.
Wall Street analysts pointed to Citrini’s research as an explanation for some of the declines. The sell-off highlights the nervousness of investors about America’s heavy spending on AI infrastructure and the impact the technology will have on the economy.
It comes as the leaders of AI companies make ever more outlandish predictions about the power of their technologies.
• Tech barons are warning of AI doom, so why don’t they slow down?
Sam Altman, chief executive of OpenAI, told the AI Impact Summit in Delhi last week that AI “superintelligence”, a theoretical milestone where machines could surpass human performance, could soon be more effective than him as a chief executive.
“AI superintelligence at some point on its development curve would be capable of doing a better job being the CEO of a major company than any executive, certainly me,” Altman said. He added: “On our current trajectory, we believe we may be only a couple of years away from early versions of true superintelligence.”
• Dominic O’Connell: AI set to cut a swathe through white-collar professions
Sectors including software and professional services have all faced sell-offs because of fears about AI-related disruption. Alarm was triggered this month by a new legal tool from Anthropic’s Claude large language model that assists with tasks across legal, sales, marketing and data analysis.
However, many analysts and investors on Monday argued the sell-off was overblown. Kyle Harrison, a venture capitalist in San Francisco, wrote on X: “Are we all stupid? Is this mass psychosis?”
Michael O’Rourke, chief market strategist at Jones Trading, told Bloomberg: “It is a remarkable reaction. I have seen this market exhibit incredible resilience in the face of actual negative news. Now a literal work of fiction sends it into a tailspin.”