{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6IG27BCKEPU3H7Y4I3QZTIFMQ4","short_pith_number":"pith:6IG27BCK","schema_version":"1.0","canonical_sha256":"f20daf844a23e9b3ff1c46e199a0ac870c91c29009f291f1a9fec0e074576ebd","source":{"kind":"arxiv","id":"2505.18003","version":1},"attestation_state":"computed","paper":{"title":"An Example Safety Case for Safeguards Against Misuse","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jonah Weinbaum, Joshua Clymer, Kimberly Mai, Robert Kirk, Selena Zhang, Xander Davies","submitted_at":"2025-05-23T15:06:21Z","abstract_excerpt":"Existing evaluations of AI misuse safeguards provide a patchwork of evidence that is often difficult to connect to real-world decisions. To bridge this gap, we describe an end-to-end argument (a \"safety case\") that misuse safeguards reduce the risk posed by an AI assistant to low levels. We first describe how a hypothetical developer red teams safeguards, estimating the effort required to evade them. Then, the developer plugs this estimate into a quantitative \"uplift model\" to determine how much barriers introduced by safeguards dissuade misuse (https://www.aimisusemodel.com/). This procedure "},"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":"2505.18003","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T15:06:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"40083988407b8443a1d0144242a24c117dd61f002f53b0d0ca9d892acf22b20f","abstract_canon_sha256":"2977b2eb22d913779950d96579a7627eebd30a0867255c280404f442185d150a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:35.100424Z","signature_b64":"uQHbNqEPnNpbJsj8uWXJFw1XMNE38uwYEIWn+TjhjGGRz19B9AYLLFepjHuEQktElNSOsMGvrhrasRxI3PPiAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f20daf844a23e9b3ff1c46e199a0ac870c91c29009f291f1a9fec0e074576ebd","last_reissued_at":"2026-07-05T11:08:35.099914Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:35.099914Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"An Example Safety Case for Safeguards Against Misuse","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jonah Weinbaum, Joshua Clymer, Kimberly Mai, Robert Kirk, Selena Zhang, Xander Davies","submitted_at":"2025-05-23T15:06:21Z","abstract_excerpt":"Existing evaluations of AI misuse safeguards provide a patchwork of evidence that is often difficult to connect to real-world decisions. To bridge this gap, we describe an end-to-end argument (a \"safety case\") that misuse safeguards reduce the risk posed by an AI assistant to low levels. We first describe how a hypothetical developer red teams safeguards, estimating the effort required to evade them. Then, the developer plugs this estimate into a quantitative \"uplift model\" to determine how much barriers introduced by safeguards dissuade misuse (https://www.aimisusemodel.com/). This procedure "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.18003","kind":"arxiv","version":1},"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/2505.18003/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":"2505.18003","created_at":"2026-07-05T11:08:35.099973+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.18003v1","created_at":"2026-07-05T11:08:35.099973+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.18003","created_at":"2026-07-05T11:08:35.099973+00:00"},{"alias_kind":"pith_short_12","alias_value":"6IG27BCKEPU3","created_at":"2026-07-05T11:08:35.099973+00:00"},{"alias_kind":"pith_short_16","alias_value":"6IG27BCKEPU3H7Y4","created_at":"2026-07-05T11:08:35.099973+00:00"},{"alias_kind":"pith_short_8","alias_value":"6IG27BCK","created_at":"2026-07-05T11:08:35.099973+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6IG27BCKEPU3H7Y4I3QZTIFMQ4","json":"https://pith.science/pith/6IG27BCKEPU3H7Y4I3QZTIFMQ4.json","graph_json":"https://pith.science/api/pith-number/6IG27BCKEPU3H7Y4I3QZTIFMQ4/graph.json","events_json":"https://pith.science/api/pith-number/6IG27BCKEPU3H7Y4I3QZTIFMQ4/events.json","paper":"https://pith.science/paper/6IG27BCK"},"agent_actions":{"view_html":"https://pith.science/pith/6IG27BCKEPU3H7Y4I3QZTIFMQ4","download_json":"https://pith.science/pith/6IG27BCKEPU3H7Y4I3QZTIFMQ4.json","view_paper":"https://pith.science/paper/6IG27BCK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.18003&json=true","fetch_graph":"https://pith.science/api/pith-number/6IG27BCKEPU3H7Y4I3QZTIFMQ4/graph.json","fetch_events":"https://pith.science/api/pith-number/6IG27BCKEPU3H7Y4I3QZTIFMQ4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6IG27BCKEPU3H7Y4I3QZTIFMQ4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6IG27BCKEPU3H7Y4I3QZTIFMQ4/action/storage_attestation","attest_author":"https://pith.science/pith/6IG27BCKEPU3H7Y4I3QZTIFMQ4/action/author_attestation","sign_citation":"https://pith.science/pith/6IG27BCKEPU3H7Y4I3QZTIFMQ4/action/citation_signature","submit_replication":"https://pith.science/pith/6IG27BCKEPU3H7Y4I3QZTIFMQ4/action/replication_record"}},"created_at":"2026-07-05T11:08:35.099973+00:00","updated_at":"2026-07-05T11:08:35.099973+00:00"}