OpenAI and Anthropic clash over AI safety in Washington manifestos

OpenAI and Anthropic, the two most influential AI labs on the planet, are waging a policy war in Washington over a question that sounds philosophical but is deeply practical: should the most powerful AI systems be opened up to everyone, or locked down behind safety gates? Two manifestos, two worldviews OpenAI fired first. On June 9, the company published a 13-page policy document titled “Industrial Policy for the Intelligence Age,” which reads like a techno-optimist’s wish list. Broad access to AI. Democratization of AI-derived wealth. Worker involvement in deployment decisions. Adaptive regulatory frameworks that keep pace with innovation rather than strangling it. Anthropic took the opposite approach. On July 8, the company updated its “Responsible Scaling Policy” to version 3.4, doubling down on safety thresholds, third-party oversight, and rigorous testing before releasing increasingly capable models. Anthropic’s co-founder Dario Amodei laid the intellectual groundwork in a June 2026 essay called “Policy on the AI Exponential,” which argued that AI is advancing faster than any regulatory body can reasonably keep up with. Over 1,100 AI engineers, including Anthropic’s CEO, publi...