The AI boom entered a new phase Monday when warnings from the industry’s own leaders translated directly into market losses. Anthropic↗ CEO Dario Amodei urged companies to slow the pace at which frontier models gain capability, citing fears that more autonomous systems could enable large-scale cyber abuse and other harms. OpenAI↗ CEO Sam Altman and xAI’s Elon Musk publicly supported the idea.
Investors immediately connected the safety debate to revenue expectations. Nasdaq 100 futures fell, SoftBank dropped sharply in Tokyo, and semiconductor shares weakened. If frontier labs stretch development timelines, that changes the expected pace of spending on GPUs, memory, networking gear and data centers.
The important shift is that AI safety is becoming an economic input rather than a separate policy conversation. Markets had treated faster AI progress as automatically positive for infrastructure suppliers. A credible slowdown changes the timing of that thesis even if long-term demand remains enormous.
There is also a strategic tension underneath the discussion: every firm and country worries that restraint by one party becomes advantage for another. For businesses using AI now, the durable lesson is to build modular, auditable workflows that can substitute models rather than depend completely on one provider.
AI-linked stocks sold off globally after Anthropic CEO Dario Amodei called for slowing frontier-model development, with Sam Altman and Elon Musk also backing the idea.