Dario Amodei says it's time to "pace the frontier" and will give third-party evaluators like METR access to Anthropic's models.
Anthropic CEO Dario Amodei laid out a plan to slow the pace of AI capability gains, arguing frontier labs need external checks before pushing further. Sam Altman has voiced similar sentiment, marking rare public alignment between two labs that have otherwise competed aggressively on model releases.
The plan includes opening Anthropic's models to independent evaluators such as METR to verify adherence to safety commitments, a concrete step beyond the usual safety rhetoric from lab leadership.
When the two most aggressive frontier labs both talk about pacing, it's either a genuine safety pivot or a coordinated signal to regulators ahead of tighter rules. Either way, enterprise buyers should expect slower model-to-model capability jumps and more scrutiny embedded in procurement and compliance conversations.
The daily signal, curated. Get it in your inbox.
Subscribe on LinkedIn →