Trump Administration Rejects AI Slowdown, Tells Labs to Self-Regulate

A growing chorus of artificial intelligence leaders—including executives from Anthropic, OpenAI, and Tesla—are calling for structured regulations and slower AI development to manage emerging safety risks. But the Trump administration's response is blunt: if you're worried about your own models, slow them down yourself.

Industry Pushback on Regulation#

Anthropic CEO Dario Amodei argued in a recent essay that Washington should create legal pathways for competing AI firms to collaborate on safety standards without violating antitrust rules. He also called for government support in establishing international coordination frameworks, even with China. OpenAI's Sam Altman and billionaire Elon Musk notably endorsed Amodei's position, signaling rare cross-company agreement on the need for more deliberate AI development.

But David Sacks, co-chair of the President's Council of Advisors on Science and Technology, dismissed the premise. He said lab leaders don't need permission to be cautious—they can simply choose restraint unilaterally, and questioned whether their concerns about safety were truly independent of competitive motives.

China Threat Dominates White House Thinking#

The Trump team's refusal to embrace AI regulation hinges on geopolitical competition. The administration views China's advances in artificial intelligence as both an economic and national security threat, and worries that regulatory measures could hand Beijing a technological advantage. President Trump has been explicit about this calculation: "We're leading China on AI," he said during a trip to Ireland, "and, frankly, I want to keep it that way because whoever wins AI, wins."

House Speaker Mike Johnson echoed the concern, warning that congressional intervention could cause the US to lose its AI edge. The administration's stance reflects deep skepticism that international coordination—especially with China—would ever materialize in practice.

Voluntary Oversight Emerging#

Rather than wait for regulation, some AI companies are moving toward self-policing measures. OpenAI's Altman committed to embedding third-party safety evaluators within AI firms, granting them employee-level access to review practices and flag problems. While Altman acknowledged these steps carry "significant costs," he framed deliberate pacing as justified by competitive responsibility, stating that "no amount of American competitive pressure should justify recklessness."