Anthropic's Responsible Scaling Policy: Version 3.0

Notes

First-party policy announcement. Slight discount from primary_doc prior (0.90) because the document is self-assessing Anthropic's own theory of change.

Evidence from this source (5)

• CDRSP v3 introduces a Frontier Safety Roadmap of publicly-declared, nonbinding go… support
  weight

  0.95

  method: primary_testimony · locator: Section: Updating our Responsible Scaling Policy, item 2

  “Our new RSP introduces a requirement to develop and publish a Frontier Safety Roadmap, which will describe our concrete plans for risk mitigations across the areas of Security, Alignment, Safeguards, and Policy.”

• CDRSP v3 requires Risk Reports published every 3-6 months describing model capabi… support
  weight

  0.90

  method: primary_testimony · locator: Section: Updating our Responsible Scaling Policy, item 3

  “Risk Reports will be published online (with some redactions) every 3-6 months... They will have unredacted or minimally-redacted access to the Risk Report and will subject our reasoning, analysis, and decision-making to a comprehensive public review.”

• CDPre-specified AI capability thresholds have proven ambiguous in practice: model… support
  weight

  0.90

  method: primary_testimony · locator: Section: Assessing our theory of change

  “We found pre-set capability levels to be far more ambiguous than we anticipated... The science of model evaluation isn't well-developed enough to provide dispositive answers.”

• CDAs of early 2026, US federal policy has shifted toward AI competitiveness and e… support
  weight

  0.75

  method: journalistic_report · locator: Section: Assessing our theory of change

  “The policy environment has shifted toward prioritizing AI competitiveness and economic growth, while safety-oriented discussions have yet to gain meaningful traction at the federal level.”

• CDRSP v3 explicitly separates Anthropic's unilateral mitigation commitments from… support
  weight

  0.95

  method: primary_testimony · locator: Section: Updating our Responsible Scaling Policy, item 1

  “Our RSP now outlines two sets of mitigations: first, the mitigations that we plan to pursue regardless of what others do; and second, an ambitious capabilities-to-mitigations map that, we believe, would help adequately manage the risks from advanced AI if implemented across the AI industry.”