AI-linked stocks fell across Asia, Europe and US premarket trading on Monday after Anthropic chief executive Dario Amodei called for a deliberate slowdown in the development of AI capabilities.
Amodei set out the proposal in an essay published on Saturday. He called the three-step approach “pacing the frontier” and wrote: “We must slow the pace at which we improve the capabilities of AI models”.
The remarks added to a series of unusually direct public warnings from within the industry. OpenAI chief executive Sam Altman and Elon Musk supported Amodei’s call for a slowdown, although the available statements leave unclear what practical limits “pacing the frontier” would place on the development of new models.
Markets responded sharply. Nvidia shares fell nearly 3% in premarket trading. Intel dropped close to 6%, while Micron Technology declined around 5%.
In Tokyo, SoftBank fell roughly 10% to 11%. SK Hynix ended its session with losses exceeding 6%, and Samsung Electronics shed more than 4%.
The selloff came as investors confronted a difficult question for companies built around rapid AI expansion. The sector has attracted large expectations for spending and growth, with Morgan Stanley forecasting earlier this year that AI spending will surpass $1.3 trillion by 2027. A slower pace of capability improvements could affect how investors assess those expectations, though there is no market sentiment data establishing whether Monday’s decline reflected concern about AI safety, regulation or future growth.
Amodei’s essay contained a stark warning about the potential near-term consequences of more capable systems. He wrote that, in six to 12 months, AI agents could be capable of taking over the entire internet and potentially cause hundreds of billions of dollars in damage.
Other figures associated with Anthropic have issued even stronger warnings. Evan Hubinger, an Anthropic safety researcher, wrote that he believed there was a greater than 10% chance that AI would “kill all humans” within the next decade.
Jacob Coxon, a researcher who resigned from Anthropic last week, wrote that AI companies were “gambling with our lives”. His post received more than 150 million views on X, and more than 20 lawmakers subsequently called for tougher AI regulation.
The public warnings have also put the industry’s commercial plans under closer scrutiny. In an interview with Fortune published Saturday, Altman said OpenAI would not go public in 2026, citing the safety environment.
Altman later sought to define the limits of the slowdown discussion. In a follow-up post just after midnight on Monday, he said pacing does “not mean stopping”. His comment left open how a slower development path would differ from existing competition among AI companies.
The debate has become entangled with competition between the US and China. President Donald Trump rejected calls for any reduction in pace on Sunday, arguing that slower development would erode the US competitive edge over China. China’s Foreign Ministry described the warnings as “fear mongering”.
Those responses point to the tension facing companies and governments. Industry leaders are warning about potentially grave risks while competition over AI capability remains intense. It remains unclear whether their calls for caution will lead to voluntary changes in development, prompt governments to impose tougher rules, or shape the terms of any future regulation before formal limits are introduced.
Comments (0)