Signals flagged in the original
Provisional estimate — refines shortly Full breakdown ↓
Anthropic has allowed external evaluators to access its systems to assess AI risks, part of its commitment to improving safety measures. The nonprofit METR is now embedded in Anthropic's office to evaluate its AI models.
Our neutral summary
Anthropic has allowed external evaluators to access its systems to assess AI risks, part of its commitment to improving safety measures. The nonprofit METR is now embedded in Anthropic's office to evaluate its AI models.
The reporting belongs to Ars Technica; we don't reproduce it.
No note attached
on this article.
Anthropic gives outside AI watchdog “employee-like” access
Anthropic provides external evaluators with access to assess AI risks