Anthropic CEO's Antitrust Waiver Plan Sparks Skepticism Amid AI Safety Concerns
Anthropic CEO Dario Amodei has proposed an antitrust waiver to allow major AI companies to coordinate on safety protocols without triggering legal liability. His proposal, called 'pacing the frontier,' aims to keep AI safety efforts in sync with rapid capability development.
The framework includes three parts: embedding independent third-party evaluators inside AI labs to monitor safety and alignment work, establishing coordinated safety standards among democratic nations, and opening dialogue channels with authoritarian states to prevent a global AI arms race.
While industry peers such as OpenAI have expressed support for the evaluator access model, they remain skeptical about the antitrust waiver. FTC Chair Andrew Ferguson has voiced concern that granting exemptions to powerful AI companies could lock in their advantages over new entrants.
The timing of Amodei's proposal is significant, given recent warnings about self-improvement capabilities in frontier models and concerns about imminent risks from AI advancements projected within the next six to twelve months.