OpenAI Wants to Know if an AI Industry Slowdown Would Even Be Legal

Details

OpenAI has been lobbying Congress in recent weeks to clarify whether orchestrating an industry-wide slowdown in frontier AI development would violate the Sherman Antitrust Act. The concern: formal safety coordination between competing labs could legally constitute output restriction, exposing companies to antitrust liability. The push follows a blog post by OpenAI chief scientist Jakub Pachocki arguing that coordinating to slow future development is essential for safety. He expects 'voluntary slowdowns to become commonplace until shared safety bars are established' — a significant public statement from one of the most senior technical figures at any frontier lab. Legal scholars say the exposure is real. Nicholas Felstead of the Australian Competition and Consumer Commission argued in March that a coordinated pause could constitute output restriction under the Sherman Antitrust Act. The precise details of any agreement would determine legality, but legal uncertainty alone 'can act as a powerful deterrent.' A bipartisan, bicameral bill — the Collaboration on Adversarial Threats and Security Risks Act — was introduced in July to explicitly authorize AI labs to coordinate on safety without antitrust risk. The House version has been referred to the Judiciary Committee but not scheduled. Caleb Knapp of the AI Policy Network says Congress has a 'growing appetite' for action but expects any legislation to wait until after midterm elections.