{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:NVAAC5K4RAHLCQSE5IBH37DI4J","short_pith_number":"pith:NVAAC5K4","schema_version":"1.0","canonical_sha256":"6d4001755c880eb14244ea027dfc68e249e417b5e7cde004e5fc52ef834fc13a","source":{"kind":"arxiv","id":"2503.11917","version":3},"attestation_state":"computed","paper":{"title":"A Framework for Evaluating Emerging Cyberattack Capabilities of AI","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Allan Dafoe, Anna Wang, Four Flynn, Lihao Liang, Mikel Rodriguez, Raluca Ada Popa","submitted_at":"2025-03-14T23:05:02Z","abstract_excerpt":"As frontier AI models become more capable, evaluating their potential to enable cyberattacks is crucial for ensuring the safe development of Artificial General Intelligence (AGI). Current cyber evaluation efforts are often ad-hoc, lacking systematic analysis of attack phases and guidance on targeted defenses. This work introduces a novel evaluation framework that addresses these limitations by: (1) examining the end-to-end attack chain, (2) identifying gaps in AI threat evaluation, and (3) helping defenders prioritize targeted mitigations and conduct AI-enabled adversary emulation for red team"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2503.11917","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-03-14T23:05:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"27d68a7ba845117a92443aeaab7f1ae350c97d58fb54344946a86e43a91c0127","abstract_canon_sha256":"c10703f4a1224efe269c86974a0757338bff3e84396cecc98b3b935d82a1b563"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:52:21.017206Z","signature_b64":"pFVIxonJzQnKqr0Ss2Iy306cvgJhQjJr4DgFMssRx6Up5ffYvIhS/9wQiIAUJWD4dmTToErtr/StHc9Twh9uAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6d4001755c880eb14244ea027dfc68e249e417b5e7cde004e5fc52ef834fc13a","last_reissued_at":"2026-07-05T10:52:21.016669Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:52:21.016669Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Framework for Evaluating Emerging Cyberattack Capabilities of AI","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Allan Dafoe, Anna Wang, Four Flynn, Lihao Liang, Mikel Rodriguez, Raluca Ada Popa","submitted_at":"2025-03-14T23:05:02Z","abstract_excerpt":"As frontier AI models become more capable, evaluating their potential to enable cyberattacks is crucial for ensuring the safe development of Artificial General Intelligence (AGI). Current cyber evaluation efforts are often ad-hoc, lacking systematic analysis of attack phases and guidance on targeted defenses. This work introduces a novel evaluation framework that addresses these limitations by: (1) examining the end-to-end attack chain, (2) identifying gaps in AI threat evaluation, and (3) helping defenders prioritize targeted mitigations and conduct AI-enabled adversary emulation for red team"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.11917","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2503.11917/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2503.11917","created_at":"2026-07-05T10:52:21.016739+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.11917v3","created_at":"2026-07-05T10:52:21.016739+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.11917","created_at":"2026-07-05T10:52:21.016739+00:00"},{"alias_kind":"pith_short_12","alias_value":"NVAAC5K4RAHL","created_at":"2026-07-05T10:52:21.016739+00:00"},{"alias_kind":"pith_short_16","alias_value":"NVAAC5K4RAHLCQSE","created_at":"2026-07-05T10:52:21.016739+00:00"},{"alias_kind":"pith_short_8","alias_value":"NVAAC5K4","created_at":"2026-07-05T10:52:21.016739+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.13079","citing_title":"The Emergence of Autonomous Penetration Capabilities in Large Language Model-Powered AI Systems","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03811","citing_title":"AI Agents Enable Adaptive Computer Worms","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13079","citing_title":"The Emergence of Autonomous Penetration Capabilities in Large Language Model-Powered AI Systems","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2507.06261","citing_title":"Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities","ref_index":71,"is_internal_anchor":false},{"citing_arxiv_id":"2509.06921","citing_title":"Neuro-Symbolic AI for Cybersecurity: State of the Art, Challenges, and Opportunities","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20389","citing_title":"CyberCertBench: Evaluating LLMs in Cybersecurity Certification Knowledge","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NVAAC5K4RAHLCQSE5IBH37DI4J","json":"https://pith.science/pith/NVAAC5K4RAHLCQSE5IBH37DI4J.json","graph_json":"https://pith.science/api/pith-number/NVAAC5K4RAHLCQSE5IBH37DI4J/graph.json","events_json":"https://pith.science/api/pith-number/NVAAC5K4RAHLCQSE5IBH37DI4J/events.json","paper":"https://pith.science/paper/NVAAC5K4"},"agent_actions":{"view_html":"https://pith.science/pith/NVAAC5K4RAHLCQSE5IBH37DI4J","download_json":"https://pith.science/pith/NVAAC5K4RAHLCQSE5IBH37DI4J.json","view_paper":"https://pith.science/paper/NVAAC5K4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.11917&json=true","fetch_graph":"https://pith.science/api/pith-number/NVAAC5K4RAHLCQSE5IBH37DI4J/graph.json","fetch_events":"https://pith.science/api/pith-number/NVAAC5K4RAHLCQSE5IBH37DI4J/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NVAAC5K4RAHLCQSE5IBH37DI4J/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NVAAC5K4RAHLCQSE5IBH37DI4J/action/storage_attestation","attest_author":"https://pith.science/pith/NVAAC5K4RAHLCQSE5IBH37DI4J/action/author_attestation","sign_citation":"https://pith.science/pith/NVAAC5K4RAHLCQSE5IBH37DI4J/action/citation_signature","submit_replication":"https://pith.science/pith/NVAAC5K4RAHLCQSE5IBH37DI4J/action/replication_record"}},"created_at":"2026-07-05T10:52:21.016739+00:00","updated_at":"2026-07-05T10:52:21.016739+00:00"}