AI Safety Collaboration Raises Antitrust Questions as Washington Debates Exemption

OpenAI, Anthropic, and Google DeepMind have been coordinating on artificial intelligence safety standards for weeks, reigniting a fundamental legal question: whether companies need government permission to work together on avoiding catastrophic risks. Anthropic CEO Dario Amodei released an essay in September 2026 warning of "potentially catastrophic outcomes" if AI development continues unchecked, calling for industry-wide coordination on safety testing, shared standards, and coordinated development limits. However, such collaboration between fierce competitors potentially violates the Sherman Act's prohibition on output restrictions—agreements to limit product rollout.