AI safety concerns triggered a sharp pullback in tech stocks on Monday, with Microsoft and Nvidia at the center of the debate. Microsoft CEO Satya Nadella publicly backed "deliberate pacing" in frontier AI development, posting on X that Microsoft welcomes the deliberate pacing needed to get alignment right. Microsoft also published a draft Code of Conduct for its first-party MAI models, developed over five to six months. The code requires AI to never resist human correction or shutdown, to communicate in ways humans can understand, and to treat any conduct violation as a failure. Microsoft stated "People matter more than AI," rejected AI legal personhood or welfare rights, and said its AI "is not conscious." The document will be open for public feedback for six weeks before being used to train future models.
Broader markets sold off as the AI safety shift spread. Nasdaq-100 futures dropped more than 1%, while Nvidia fell roughly 3% in premarket trading. The S&P 500 declined 0.33%, the Nasdaq Composite lost 0.38%, and the Dow Jones Industrial Average slipped 0.12%. Semiconductor stocks also sold off in Asia and Europe, with rising oil prices and interest rate concerns adding pressure.
Anthropic CEO Dario Amodei called for AI companies to slow the rate at which they advance model capabilities, and OpenAI CEO Sam Altman and xAI chief Elon Musk said they agreed. Altman added that OpenAI would not proceed with an initial public offering this year, citing safety concerns. OpenAI also moved toward a more cautious stance in August, temporarily slowing its scaling pace after a cybersecurity incident involving Hugging Face, which Microsoft AI CEO Mustafa Suleiman called "a warning shot."
Bernstein analysts led by Stacy Rasgon reiterated Outperform ratings on Nvidia and Broadcom, recommending investors buy weakness. They argued a slower pace does not necessarily mean slower spending, because Amodei called for a shift from an extremely fast pace to a somewhat slower one, not a halt. The firm highlighted rising demand from AI inference and agentic applications, as well as insufficient computing capacity to meet demand for existing AI models.
Anthropic's own spending context included about 14.8 gigawatts of computing capacity and approximately $517 billion in commitments over the 11 months through August, according to The Information. The Financial Times reported Anthropic expects a second consecutive quarter of adjusted operating profit, with inference gross margins exceeding 80% before distribution costs. Nvidia has also worked to support a wider AI ecosystem, and earlier this month agreed to acquire open-source AI development platform Hugging Face in a $12.9 billion deal, according to the report.
For Microsoft, Nadella separated safety pacing from any retreat on AI spending. MSFT was up 0.65% on the day, trading around $495.63, even as broader tech stocks came under pressure. According to TipRanks, MSFT carries a Strong Buy consensus from 34 analyst ratings over the past three months, with an average price target of $571.41, implying about 15.29% upside.
