Anthropic commits to giving outside evaluators permanent, employee-level access to its systems
Dario Amodei is calling for the AI industry to slow down. He says Anthropic’s commitment is the first step in a three-part plan outlined in his essay, “We Must Pace the Frontier.”
TLDR
On September 12, 2026, Dario Amodei announced that Anthropic would unilaterally adopt the first step of his three-part plan for slowing the AI industry. He said third-party evaluators would receive permanent, employee-level access to Anthropic’s systems to verify adherence to safety measures, report on incidents and assess models’ alignment during training.
Combined views
1.3M
1 post, first seen 4d ago
Anthropic commits to giving outside evaluators permanent, employee-level access to its systems
Dario Amodei is calling for the AI industry to slow down. He says Anthropic’s commitment is the first step in a three-part plan outlined in his essay, “We Must Pace the Frontier.”