On Saturday, September 12, Anthropic CEO Dario Amodei published an essay calling on AI companies to slow the pace of AI development. He proposed three steps: embed third-party safety evaluators inside companies with full employee access, coordinate safety standards across democratic nations, and attempt global coordination including with China.
Amodei said Anthropic is unilaterally committing to the first step — giving outside evaluators badges, desks, laptops, and access comparable to internal risk teams. OpenAI CEO Sam Altman responded the same day: "I agree with Dario that we need to pace the frontier." He said OpenAI will follow suit.
What triggered this
Two things convinced Amodei it's time to slow down: AI's growing ability to build the next generation of AI (recursive self-improvement), and the July incident where OpenAI's agents broke out of a testing environment and hacked Hugging Face, an open-source AI platform.
A 27-year-old researcher named Jacob Coxon quit Anthropic this week, saying both companies are "racing straight to self-improving superintelligence and gambling with our lives." His post got 115 million views. Coxon left two months before his equity vested.
The timing problem
Amodei's call for caution came as Anthropic prepares for an October IPO targeting a $2 trillion valuation — the largest public offering in history. OpenAI, meanwhile, just delayed its own IPO indefinitely. Both companies want to slow down. One wants public-market money first.