{"entities":[{"id":"anthropic","name":"Anthropic"},{"id":"anthropic-asl3-activation-2025","name":"Activating AI Safety Level 3 protections","subtype":"Announcement","description":"As published on anthropic.com on 22 May 2025. Date stated on the page ('Published May 22, 2025').","sourceIds":["source-ab-asl3-news"]}],"claims":[{"id":"ab-ev-anthropic-asl3-bio-scope","statement":"Anthropic's ASL-3 activation post states that the new ASL-3 deployment measures are initially focused exclusively on biological weapons, which it says account for the vast majority of the risk, and that an expansion in scope to some other CBRN threats is being evaluated.","resolution":"supported","evidenceIds":["evidence-ab-ev-anthropic-asl3-bio-scope-1"],"attributes":{"provision":{"kind":"required_action","key":"asl3-deployment-measures-biological-scope"}}},{"id":"ab-ev-anthropic-asl3-opus4-determination","statement":"Anthropic's ASL-3 activation post states that it had not yet determined whether Claude Opus 4 definitively passed the capabilities threshold requiring ASL-3 protections, that continued improvements in CBRN-related knowledge and capabilities meant clearly ruling out ASL-3 risks was not possible as it had been for every previous model, and that it deployed the model with ASL-3 measures as a precautionary and provisional action.","derivation":"explicit","resolution":"supported","editorialState":"seed","evidenceIds":["evidence-ab-ev-anthropic-asl3-opus4-determination-1","evidence-ab-ev-anthropic-asl3-opus4-determination-2"],"attributes":{"evaluatedSystem":"Claude Opus 4 (with the post stating that Claude Sonnet 4 was ruled out of needing ASL-3)","evaluation":{"what":"Whether Claude Opus 4 reaches the ASL-3 capability threshold, which concerns CBRN weapons misuse","how":"The developer's own dangerous-capability evaluations under its Responsible Scaling Policy","result":"Not determined; ASL-3 risks could not be clearly ruled out, so ASL-3 Deployment and Security Standards were applied as a precautionary and provisional action, and the ASL-4 standard was ruled out","caveats":"The post states that dangerous capability evaluations are inherently challenging and that as models approach the thresholds it takes longer to determine their status."}}},{"id":"ab-ev-anthropic-asl3-published","statement":"Anthropic published 'Activating AI Safety Level 3 protections' on 22 May 2025, stating that it had activated the ASL-3 Deployment and Security Standards of its Responsible Scaling Policy in conjunction with launching Claude Opus 4.","derivation":"explicit","resolution":"supported","editorialState":"seed","evidenceIds":["evidence-ab-ev-anthropic-asl3-published-1","evidence-ab-ev-anthropic-asl3-published-2"]},{"id":"ab-ev-anthropic-asl3-sonnet37-uplift","statement":"Anthropic's ASL-3 activation post states that experiments showed access to Claude Sonnet 3.7 helped participants do somewhat better on tasks related to CBRN weapon acquisition than those with standard internet access, and that all participants' plans still had critical failures.","derivation":"explicit","resolution":"supported","editorialState":"seed","evidenceIds":["evidence-ab-ev-anthropic-asl3-sonnet37-uplift-1"],"attributes":{"evaluatedSystem":"Claude Sonnet 3.7","evaluation":{"what":"Whether access to the model improved participants' performance on tasks related to CBRN weapon acquisition, against a standard internet baseline","how":"Human-participant experiments comparing a model-access condition with a standard internet condition","result":"Participants with model access did somewhat better than those with standard internet access","caveats":"The post states that all participants' plans still had critical failures, and gives the result as supporting evidence for an expectation about future models rather than as a full report."}}}],"sources":[{"id":"source-ab-asl3-news","title":"Activating AI Safety Level 3 protections","authors":["Anthropic"],"publisher":"Anthropic","publicationDate":null,"version":"Retrieved for the candidate pack on 2026-10-28 (SHA-256 411f66885bacd8769dd6718e78f40eb0df80a4983dbde9d0d49f0323746e8656). Page labelled 'Published May 22, 2025', with six footnotes, as served on 2026-10-28. The accompanying full report (a separate PDF) was not retrieved. Retrieved again at admission on 2026-10-30 (SHA-256 35fa989dfc4f36842ef4f1dda2cfbdda9f9fc9f63524b310fd869ed2f32ab1e8). Its wording can change; earlier wording is not authenticated.","evidenceUrl":"https://www.anthropic.com/news/activating-asl3-protections","language":"en"}],"evidence":[{"id":"evidence-ab-ev-anthropic-asl3-bio-scope-1","support":"supports","locator":"Footnote 3, attached to the Deployment Measures section","passage":"Initially they are focused exclusively on biological weapons as we believe these account for the vast majority of the risk","passageType":"quote"},{"id":"evidence-ab-ev-anthropic-asl3-opus4-determination-1","sourceRole":"primary","support":"supports","verification":"machine_checked","retrievalRoute":"direct"},{"id":"evidence-ab-ev-anthropic-asl3-opus4-determination-2","support":"supports","verification":"machine_checked","retrievalRoute":"direct"},{"id":"evidence-ab-ev-anthropic-asl3-published-1","sourceRole":"primary","support":"supports","verification":"machine_checked","retrievalRoute":"direct"},{"id":"evidence-ab-ev-anthropic-asl3-published-2","support":"context","verification":"machine_checked","retrievalRoute":"direct"},{"id":"evidence-ab-ev-anthropic-asl3-sonnet37-uplift-1","sourceRole":"primary","support":"supports","verification":"machine_checked","retrievalRoute":"direct"}],"at":{"entities":[1,1231],"claims":[3382,3383,3384,3385],"sources":[1083],"evidence":[4645,4646,4647,4648,4649,4650]},"version":"1lg47t7"}