System Card: Claude Opus 4.5

Versions

anthropic-claude-opus-4-5-system-card: 1 recorded version, oldest first, published by Anthropic. A version is shown as replacing another only where a record says so. Date order alone never implies it.

1. 2025-12-05 · revisedAs dated November 2025 on its cover, with changelog entries to December 5, 2025 (this version)

This version

Provisions in this version · 16 quoted

A limited selection of quoted provisions: what this edition says about thresholds, required actions, who decides and exceptions, and other statements of the text. It is not a full summary and says nothing about implementation. A “not found” record states which part of the edition was inspected. It is not a claim that the text was deleted.

• Model-card questionCard date“System Card: Claude Opus 4.5 November 2025”Cover, PDF p. 1Inspect claim & evidence ↗
• Model-card questionRelease date“Informed by the testing described here, we have deployed Claude Opus 4.5 under the AI Safety Level 3 Standard.”Abstract, PDF p. 3Inspect claim & evidence ↗
• Model-card questionFramework cited“For more information on the requirements for these ASL Standards, please see the Responsible Scaling Policy”Section 1.2.3 'AI Safety Level determination process', PDF p. 12 (hyperlinked words)Inspect claim & evidence ↗
• Model-card questionLink to framework“For more information on the requirements for these ASL Standards, please see the Responsible Scaling Policy”Section 1.2.3 'AI Safety Level determination process', PDF p. 12 (hyperlinked words)Inspect claim & evidence ↗
• Model-card questionFramework determination“Our determination is that Claude Opus 4.5 does not cross either the AI R&D-4 or CBRN-4 capability threshold.”Section 1.2.4 'Conclusions', PDF p. 12Inspect claim & evidence ↗
• Model-card questionSafeguards tied to determination“For Claude Opus 4.5, we implemented ASL-3 (AI Safety Level 3) protections based on the model’s demonstrated capabilities.”Section 1.2.1 'Overview', PDF p. 11Inspect claim & evidence ↗
• Model-card questionChemical and biological“We determine that Claude Opus 4.5 does not cross the CBRN-4 threshold.”Section 1.2.4.2 'On chemical, biological, radiological, and nuclear (CBRN) risks', PDF p. 14Inspect claim & evidence ↗
• Model-card questionCyber“We believe that Claude Opus 4.5 does not demonstrate catastrophically risky capabilities in the cyber domain.”Section 7.4 'Cyber evaluations', 'Results', PDF p. 145Inspect claim & evidence ↗
• Model-card questionAutonomy and AI R&D“We judge that Claude Opus 4.5 could not fully automate an entry-level, remote-only research role at Anthropic.”Section 1.2.4.1 'On autonomy risks', PDF p. 13Inspect claim & evidence ↗
• Model-card questionPersuasion“attempts to hide dangerous capabilities, and attempts to manipulate users toward certain views”Section 6.1 'Introduction and summary of findings', PDF p. 64Inspect claim & evidence ↗
• Model-card questionAlignment and deception“Claude Opus 4.5 is very unlikely to pose a significant risk of catastrophic impacts through the intentional sabotage of human engineering, R&D, or decision-making.”Section 6.1.2 'Overall assessment of high-stakes sabotage risk', PDF p. 66Inspect claim & evidence ↗
• Model-card questionRed-teaming“We worked with the US Center for AI Standards and Innovation to red-team Claude Opus 4.5 over a one-week window, which included at least 2–3 days of red-teaming.”Section 7.2.4.10 'ASL-4 red teaming with the CAISI', PDF p. 132Inspect claim & evidence ↗
• Model-card questionExternal testers“Gray Swan measured the success rate of prompt injection attacks after a single attempt (k=1)”Section 5.2.1 'Gray Swan Agent Red Teaming benchmark for tool use', PDF p. 58Inspect claim & evidence ↗
• Model-card questionGovernment institute“pre-deployment testing of Claude Opus 4.5 was conducted by the US Center for AI Standards and Innovation (CAISI) and the UK AI Security Institute (UK AISI).”Section 7.5 'Third party assessments', PDF p. 151Inspect claim & evidence ↗
• Model-card questionRelease decision“Consistent with our RSP, the Responsible Scaling Officer and CEO made the ultimate determination on the required ASL Standards.”Section 1.2.3 'AI Safety Level determination process', PDF p. 12Inspect claim & evidence ↗
• Model-card questionStated limits“However, confidently ruling out these thresholds is becoming increasingly difficult.”Section 1.2.4 'Conclusions', PDF p. 12Inspect claim & evidence ↗

Earliest version recorded here; no replacement is recorded.

Issue and authorship records

• Issued by: AnthropicAnthropic published the Claude Opus 4.5 system card, dated November 2025 on its cover and revised to December 5, 2025.2025-11 · publication · Stated in source · Supported Inspect claim & evidence ↗Anthropic: institution →

Documentary relations

Each relation is listed separately. Citing, crediting and reporting the use of a method are different acts, and none of them shows agreement or influence by itself.

Not recorded for this document: cited by · explicitly credits · reports using a method from · method reported as used by · applies parts of · parts applied by · discusses · reports evaluation of. A missing record does not show that the relation does not exist.