White House Retreats on FDA-Style AI Approval After Industry Pushback

The White House is attempting to defuse a regulatory controversy after a senior economic official suggested artificial intelligence models might require pre-release approval akin to pharmaceutical drug testing, prompting immediate resistance from technology executives and industry groups.

White House Chief of Staff Susie Wiles moved to clarify the administration's position following remarks by Kevin Hassett, director of the National Economic Council, who told Fox Business the government was considering an executive order establishing a safety testing regime similar to the Food and Drug Administration's approval process for frontier AI models. The episode has exposed a tug-of-war within the administration over how aggressively to regulate advanced AI systems.

The proposed framework would have represented a dramatic escalation in federal oversight, requiring companies to submit models for government review before public deployment. Industry-aligned policy groups raised concerns that such a regime could stifle innovation and hand competitive advantage to jurisdictions with lighter regulatory touch, particularly as China narrows the performance gap with U.S. models despite vastly lower private investment.

(The internal disagreement reflects broader tensions between economic advisers seeking to maintain American AI leadership through permissive policies and national security officials concerned about risks from increasingly capable systems. The administration has yet to articulate a unified regulatory philosophy.)

The debate unfolds as Stanford's latest Artificial Intelligence Index Report documents China's rapid progress in model capabilities. Russell Wald, executive director of the Stanford Institute for Human-Centered Artificial Intelligence, noted that despite the United States holding approximately $285 billion in private AI investment compared to China's $12 billion, Chinese developers have nearly closed the performance gap through robust open-source ecosystems and broad talent diffusion. Some analysts attribute the gains to distillation techniques that extract capabilities from existing models.

The regulatory uncertainty comes as AI transitions from technology story to geopolitical flashpoint, with implications for workforce transformation, military applications, and supply chain security increasingly dominating policy discussions at forums including the Special Competitive Studies Project's AI+ Expo.

The administration's struggle to define a coherent regulatory posture mirrors earlier debates over internet governance and biotechnology oversight, where competing imperatives of innovation, security, and public safety have historically produced fractured policy outcomes. The current impasse suggests any comprehensive federal AI framework remains months away, leaving companies to navigate a patchwork of voluntary commitments and sector-specific rules.