METR: Anthropic's AI speeds up in-house R&D ~2x; Model 4 self-reported at 12x
maksym_andr · x · 2026-11-06
METR published an independent assessment of Anthropic's October 2026 AI R&D acceleration:
- Interviews and surveys point to a wide range of estimates centered on 2x acceleration in research areas attributable to AI
- Anthropic's informal Slack polls show self-reported uplift climbing steadily: Mythos 5 at 5x, latest Model 4 at 12x (METR treats these as very loose evidence)
- Individual uplift doesn't directly translate to overall acceleration, which also depends on labor input growth
- Anthropic redacted most of the most informative evidence, citing risk of accelerating other AI developers — making public judgment difficult
A rare third-party look inside a frontier lab's AI-driven research acceleration and its risk implications.
More from Safety
- UK cyber bill adds 'loss of oversight or control of an AI agent' clause, still undefined — mclynd · 2026-11-06
- Dev warns: don't link your bank account to AI companies — arthurcolle · 2026-11-06
- Debate: do confidentiality deals justify staying silent on AI timelines? — albrgr · 2026-11-06
- Microsoft: threat actors use agentic AI tool ARTEX to harvest exposed secrets from GitHub — davemccollough · 2026-11-06
- Researchers clash over Anthropic report: conclusions rest on subjective staff assessments — danielrock · 2026-11-06
- Dimitris Papail: safety researcher may be the last technical AI job left — DimitrisPapail · 2026-11-06