OpenAI Asks Congress if AI Development Slowdown Violates Antitrust Law

OpenAI has approached members of Congress in recent weeks requesting clear guidance on a potentially precedent-setting question: Would coordinating an industry-wide slowdown of frontier AI development violate antitrust law?

According to people close to the company speaking with WIRED, AI leaders are concerned that existing antitrust regulations could legally prevent the kind of coordinated pause on advanced AI research that some safety advocates have called for.

The antitrust dilemma

The inquiry highlights a tension at the heart of AI governance. While safety researchers and some policymakers have argued that leading AI companies should slow or coordinate their development timelines to address safety concerns, such coordination among competitors typically raises red flags under U.S. antitrust law.

Antitrust regulations are designed to prevent competitors from colluding in ways that could harm consumers or stifle competition. An agreement among major AI companies to simultaneously slow development could theoretically fall into this category, even if the stated motivation is safety rather than market manipulation.

The question becomes particularly complex in the AI sector, where the leading companies—OpenAI, Anthropic, Google DeepMind, and others—are racing to develop increasingly capable systems. Any coordinated slowdown would require buy-in from multiple competitors, potentially creating the appearance of collusion.

Why it matters

This legal uncertainty could become a significant obstacle to implementing AI safety proposals that depend on industry coordination. If antitrust law prevents companies from jointly agreeing to development pauses or safety protocols, policymakers may need to create new regulatory frameworks that explicitly authorize such coordination—or find alternative approaches to AI governance that don't require competitor cooperation.

The broader safety context

OpenAI's inquiry comes as the company has faced increased scrutiny over its safety practices. The company has recently overhauled its safety protocols and acknowledged issues with its approach to preventing AI agents from going rogue, according to previous WIRED reporting.

The question of whether to slow AI development has become increasingly urgent among researchers concerned about the pace of capability gains in frontier models. However, translating those concerns into coordinated industry action faces not just technical and competitive challenges, but potentially legal ones as well.

The details of OpenAI's congressional outreach were first reported by WIRED.