Critique of Anthropic and OpenAI Safety Framework Replacements

The author argues that replacing Anthropic's Responsible Scaling Policy and OpenAI's Preparedness Framework with weaker, essentially commitmentless documents is deeply disappointing. The author notes that even if those previous documents had been binding, the penalties for violating them would have been minuscule for multi-trillion-dollar companies. Furthermore, the companies are only legally bound to their Frontier Compliance and Frontier Governance frameworks respectively. Open article →