An analysis by METR determined that capability growth on Anthropic's internal Epoch Capabilities Index doubled between October 2025 and October 2026 compared to the previous year. Although the organization considers automated R&D the primary driver of this sustained momentum, most supporting evidence remains redacted. Theoretical models show that maintaining this velocity demands massive productivity boosts because capability growth also depends on hardware factors like compute. Consequently, human researchers would require a four- to nine-fold increase in output through escalating automation to keep pace.
Key points
Achieving a doubling of overall capability growth requires a four- to nine-fold jump in r…: Achieving a doubling of overall capability growth requires a four- to nine-fold jump in researcher labor productivity.
Software automation must shoulder an increasing burden because physical compute constrain…: Software automation must shoulder an increasing burden because physical compute constraints continue to limit raw hardware scaling.
Sustaining this accelerated pace of innovation will necessitate continuously expanding as…: Sustaining this accelerated pace of innovation will necessitate continuously expanding assistance from artificial intelligence over time.
Sources checked by our analysis pipeline: 1
This post discusses a public METR report on AI R&D acceleration at Anthropic. I did not work on this project or see the unredacted report.1 All opinions are my own and do not represent METR or the report authors.
Thanks to Ryan Greenblatt for discussion, feedback, and help with modeling, and Daniel Kokotajlo, Eli Lifland, and Hjalmar Wjik for comments.
Earlier this week, Anthropic stated in their S1 that they “may imminently meet” the threshold of “dramatically accelerated progress due to AI” as defined in their Responsible Scaling Policy. In their latest RSP (effective July 2026), say this threshold will be met when:
They observe or expect overall capabilities progress to be going twice as fast as the fastest sustained rate before AI was contributing much to AI R&D.2
This is plausibly attributable to AI automating AI R&D, rather than acceleration in compute or other inputs.3
For the last few months, some of my colleagues at METR have been working on a deep-access investigation to understand how much Anthropic is being accelerated by its AI systems. From my reading of…
The artificial intelligence company is accelerating its path to the public markets with an imminent November debut. Reaching the projected capital targets would secure one of the largest market debuts in corporate history and cement an unprecedented baseline for AI industry valuations.
Government agencies handle critical citizen data under strict legal frameworks like GDPR and FedRAMP, making unmonitored artificial intelligence adoption a major legal and budgetary risk. Centralized platforms now allow public institutions to apply unified controls over external models and coding agents rather than managing them piecemeal. This approach safeguards public funds and ensures requests do not route through unapproved foreign infrastructure.
Developers can now operate Claude Code within remote environments powered by Amazon Bedrock AgentCore. This capability introduces dedicated security controls designed to safeguard remote sessions during execution. The enhancement ensures that automated coding tasks remain protected while functioning across distributed cloud infrastructure.