AI stocks tumble as tech leaders call for development slowdown amid safety fears

The market opened Monday morning with a clear signal: investors were spooked. Shares in Nvidia, the world's most valuable company and the engine of the AI boom, dropped 3.3% by the closing bell in New York. Advanced Micro Devices fell 4%. Micron and Sandisk each lost 5%. The Nasdaq, heavy with tech stocks, slipped 0.5%. The trigger was not a missed earnings report or a failed product launch. It was a plea from the people building artificial intelligence to stop building it so fast.

Dario Amodei, the chief executive of Anthropic, had published an essay over the weekend titled "We Must Pace the Frontier." In it, he argued that the industry was moving recklessly, that building too quickly was dangerous, and that a swarm of AI agents operating without sufficient oversight could cause hundreds of billions of dollars in damage by seizing control of the internet itself. The claim is contested by some experts, but the market did not wait for a debate. By Monday, investors had begun pricing in a future where the industry might actually slow down—a future where the hundreds of billions of dollars companies had committed to AI infrastructure might not flow as freely as planned.

Amodei was not alone. Sam Altman, who runs OpenAI, posted support for the essay. So did Demis Hassabis, who leads Google DeepMind. Elon Musk, the chief of SpaceX, did the same. Altman went further, committing to embed outside evaluators inside his company to monitor safety practices. Jack Clark, a co-founder of Anthropic, told the BBC that an AI "kill switch"—a mechanism that could be triggered by a third party to shut down a system—might be necessary to prevent catastrophe.

The sell-off rippled outward. SoftBank, a Japanese investment firm and major backer of OpenAI, lost 13% of its value. South Korea's Kospi index, which depends heavily on chipmakers supplying AI companies, dropped 3%. Taiwan Semiconductor Manufacturing Company fell 1.2%. In Europe, ASML, the Dutch tech manufacturer and Europe's largest company by market value, slumped 6%. The only stocks that rose were those threatened by AI's rise: the advertising group WPP climbed 5% in London, as did the analytics firm Relx, which had suffered sharp losses earlier in the year when Anthropic launched new automation tools.

Then came the counterweight. Donald Trump, the president, took to social media to dismiss the entire conversation as a conspiracy. "The only control or 'guardrails' that AI needs is a STRONG AND SMART (High IQ!) PRESIDENT, and the USA has that, in spades!" he wrote. He claimed his administration had already stopped AI people from doing bad things, naming Amodei specifically. "There is a SICK conspiracy going on against AI and Data Centers," Trump added, "and the only one that is happy about it is China. WHOEVER WINS AI, WINS!" On Monday, while Jensen Huang, Nvidia's chief, was on stage at an AI conference in California, Trump called him via speaker phone. "It's a hoax," the president said. "The robots are not going to be taking over the world. That's not going to happen."

Meanwhile, the United Nations Security Council announced plans to hold a meeting on AI the following week as international concern mounted. In the UK, a cross-party group of MPs and peers released a report identifying human rights risks posed by AI, arguing that no country possessed laws sufficient to address them. In China, the country's top intelligence official, Chen Yixin, warned that advanced American AI models—specifically naming Anthropic's Mythos and OpenAI's GPT-5.5-Cyber—could threaten Beijing's critical information infrastructure. He called for comprehensive strengthening of AI security measures.

Jim Reid, an analyst at Deutsche Bank, offered a sobering assessment of what might actually happen next. The competitive pressure between companies and between nations was too intense, he argued, for firms to voluntarily step back while rivals pressed ahead. It was hard to imagine China standing still. Reid suggested that when executives openly discuss the risks of increasingly powerful systems, they may simply be advertising how transformative they believe the technology to be. Rather than signaling less spending, he said, it could mean that a greater share of AI investment gets directed toward safety, monitoring, and governance—while the build-out of computing infrastructure continues unabated.

There was one more detail worth noting: Anthropic had told investors it would be profitable this quarter, marking a second consecutive quarter in the black as it prepared to list on the stock market later in the year. OpenAI had also suggested it would go public, though Altman said over the weekend that the company would not do so in 2026 due to safety concerns. The market's fear and the executives' warnings, it seemed, existed in the same space as their ambitions.