Third-party dispatches, not graded receipts
METR: Anthropic's AI R&D acceleration is roughly twice its 2024-2025 rate
AI safety organization METR published an independent assessment finding Anthropic is likely improving model capabilities on well-specified tasks at around twice the rate it did between March 2024 and October 2025, attributing this to extensive use of AI for R&D.
“Among other findings, METR says that Anthropic is “likely improving the capabilities of their models on well-specified tasks at around twice the rate they did between March 2024 and October 2025,” “likely” thanks to the “extensive” use of AI for R&D.”
Bears on: superhuman-coder-2027 · metaculus-weakly-general-ai-2028