OpenAI Wants to Know: Would an AI Industry Slowdown Even Be Legal?
AI leaders worry antitrust law could stand in the way of what they view as an increasingly urgent push to coordinate a slowdown in AI development.
By Maxwell Zeff | Business | September 10, 2026
Overview
OpenAI is exploring a question that could shape the future of AI governance: if the world’s leading AI labs decided to deliberately slow down frontier development for safety reasons, would such coordination violate antitrust law? The company has reportedly sought legal guidance on whether a coordinated slowdown among competitors might run afoul of competition regulations, reflecting growing concern among AI leaders that existing laws could block efforts to manage existential risks.
The Antitrust Dilemma
Antitrust law is designed to prevent collusion that harms competition and consumers. In most industries, competitors agreeing to limit output, divide markets, or suppress innovation would be treated as per se illegal. But AI leaders argue that an industry-wide slowdown—if motivated by safety rather than profit—could be uniquely justified. The challenge is that current antitrust frameworks make no exception for “benevolent” coordination, even when the goal is to prevent catastrophic outcomes.
OpenAI’s inquiry signals that the company is thinking seriously about a scenario in which leading labs might need to act collectively. It also highlights a tension: the same laws that protect consumers from monopolistic behavior could, in theory, make it harder for the industry to self-regulate on safety.
Why AI Leaders Are Worried
In recent months, prominent figures—including OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei—have warned that unbridled AI development could pose serious risks. Some have called for an international pause or at least a coordinated approach to safety. Yet without clear legal cover, any collective slowdown could invite lawsuits or regulatory action. The fear is that antitrust enforcement could be used to block precisely the kind of cooperation that safety advocates deem necessary.
Adding to the complexity, the global nature of AI development means that any slowdown would need to involve labs across multiple jurisdictions. Even if US law allowed it, foreign competitors might not follow suit, potentially leaving safety-focused companies at a disadvantage.
Possible Paths Forward
Legal experts suggest a few options. One is for governments to explicitly carve out safety-related coordination from antitrust enforcement, perhaps through new legislation or regulatory guidance. Another is for companies to pursue voluntary, transparent commitments that stop short of market allocation. A third is international treaty-making, though that remains slow and uncertain.
OpenAI’s legal inquiry may be an early step toward clarifying what is permissible. As AI capabilities advance, the question of whether—and how—the industry can legally slow down is likely to become more urgent. For now, the answer remains unclear, but the fact that it is being asked at all underscores how seriously some leaders are taking the risks.
What to Watch
Whether antitrust authorities in the US and abroad issue guidance on AI safety coordination. Whether OpenAI or other labs publicly endorse specific legal changes. And whether the industry’s competitive dynamics make any coordinated slowdown practical, even if it is legal.
via Wired AI