METR
Independent AI evaluation nonprofit (Model Evaluation and Threat Research) that measures autonomous capabilities and, in November 2026, published an access-based audit of AI R&D acceleration at Anthropic.
METR (Model Evaluation and Threat Research) is an independent nonprofit that evaluates frontier models for dangerous autonomous capabilities, including how far AI is speeding up AI research itself.
Anthropic R&D acceleration assessment (Nov 2026)
On Nov 5 2026 METR published a roughly 100-page assessment of Anthropic, based on privileged access from Aug 12 to Oct 29 2026 (about 2.5 full-time staff, interviews and four unreleased models). Its central estimate is that Anthropic’s internal capability benchmark is now growing about 1.9x faster than before October 2025 (95% CI 1.5-2.8x), and it guesses that perhaps 80% of that speed-up is due to AI rather than extra compute. It found no convincing bottleneck that would stop the trend over the next 12 months. METR says Anthropic, using redaction rights in the agreement, removed most of the evidence METR considered most informative; Anthropic says the cuts protect competitively sensitive information. The report lands days after Anthropic’s own S-1 filings warned it may soon reach its internal threshold for dramatically accelerated AI-assisted research.
A similar arrangement started at OpenAI in Nov 2026, where Apollo Research embedded six evaluators for four months.
Timeline
Every dated event in the wiki that involves METR, newest first.
- Scott Wiener wins Pelosi's seat, pledging federal AI regulation
The author of California's SB 1047 and SB 53 told MIT Technology Review Congress has not acted fast enough and cited the METR report on Anthropic as a sign time is short; he backs a strong federal regulator. His campaign got over $3M from Public First Action, a pro-regulation group substantially funded by Anthropic.
Sources: MIT Technology Review, “Nancy Pelosi's replacement is ready to regulate AI”
Access-based audit (Aug 12-Oct 29): Anthropic's internal capability benchmark is growing ~1.9x faster than before October 2025 (95% CI 1.5-2.8x), perhaps ~80% due to AI, with no convincing bottleneck in the next year. METR says Anthropic redacted most of the evidence it found most informative; Anthropic cites competitive confidentiality.
Connections
METR and everything one link away. Drag to move, scroll to zoom, click a node to open it.Open in the full graph →
What links here
AI lab Anthropic
Organization Apollo Research
Incident OpenAI–Hugging Face incident
Person Scott Wiener
Cited by
Cited as a source by 1 timeline events.
1 articles cite METR as a source: Anthropic