METR releases independent evaluation on Anthropic's AI R&D acceleration, showing model capability growth rate nearly doubled

Evaluation organization METR has released an independent assessment report on Anthropic, noting that as of October 2026, the growth rate of Anthropic's baseline model capabilities reached roughly 1.9 times that of previous periods. The report confirms that this acceleration was not driven by scaling training compute, but primarily by the widespread adoption of AI tools throughout the R&D pipeline, with no evident compute or data bottlenecks anticipated over the next 12 months. Anthropic exercised its redaction rights, blacking out most critical core data in the report to prevent technology spillovers.