Trump and AI CEOs clash over calls to slow advanced AI

This cluster crystallizes a turning point in U.S. AI politics. For the first time, the CEOs of Anthropic, OpenAI and xAI are publicly aligned in asking Washington to slow frontier systems and embed independent safety oversight, while the sitting U.S. president openly rejects AI risk as a hoax and a “sick conspiracy” against data centers and growth. That divergence between industry and the White House will shape how fast the largest American labs can move and what guardrails they face domestically.

Strategically, the episode exposes a rift inside the AI ecosystem: frontier labs that now see existential and cyber risks as business-critical, versus political leaders who view any slowdown as a threat to U.S. competitiveness against China. Markets reacted immediately, with AI-linked chip stocks selling off on fears that the sector may itself tap the brakes. If Trump’s stance holds and Congress remains gridlocked, it increases the odds that soft law, lab self-governance and foreign regimes will define near‑term safety norms.

For the race to AGI, this tension matters because it determines whether safety efforts run ahead of, or behind, capability scaling. A permissive U.S. policy line combined with executives asking for restraint could push safety debates into corporate boardrooms, shareholder meetings and state-level law, rather than a single federal framework.