{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:FYU3FTRQPEO3YTAF6CGKWFQNTQ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"680fe66401d26ec029696164053ea150fa9c1a99b12cef8b79e8ddbaa5ed3859","cross_cats_sorted":["cs.CR","cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-01-28T21:52:15Z","title_canon_sha256":"8d07e8671a82fba664665b6085f921bb284b2ffc041757518114103345aa1629"},"schema_version":"1.0","source":{"id":"2501.17315","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.17315","created_at":"2026-07-05T10:06:51Z"},{"alias_kind":"arxiv_version","alias_value":"2501.17315v1","created_at":"2026-07-05T10:06:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.17315","created_at":"2026-07-05T10:06:51Z"},{"alias_kind":"pith_short_12","alias_value":"FYU3FTRQPEO3","created_at":"2026-07-05T10:06:51Z"},{"alias_kind":"pith_short_16","alias_value":"FYU3FTRQPEO3YTAF","created_at":"2026-07-05T10:06:51Z"},{"alias_kind":"pith_short_8","alias_value":"FYU3FTRQ","created_at":"2026-07-05T10:06:51Z"}],"graph_snapshots":[{"event_id":"sha256:737809d1ac6c7bffa423ae297f27a2e90821050c0aace9a59281588f358ba077","target":"graph","created_at":"2026-07-05T10:06:51Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2501.17315/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As LLM agents gain a greater capacity to cause harm, AI developers might increasingly rely on control measures such as monitoring to justify that they are safe. We sketch how developers could construct a \"control safety case\", which is a structured argument that models are incapable of subverting control measures in order to cause unacceptable outcomes. As a case study, we sketch an argument that a hypothetical LLM agent deployed internally at an AI company won't exfiltrate sensitive information. The sketch relies on evidence from a \"control evaluation,\"' where a red team deliberately designs ","authors_text":"Benjamin Hilton, Buck Shlegeris, Geoffrey Irving, Joshua Clymer, Tomek Korbak","cross_cats":["cs.CR","cs.SE"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-01-28T21:52:15Z","title":"A sketch of an AI control safety case"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.17315","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:d79e9ca8720889329e41300668e9d1710b74dd57b9aaa7054a54761f83327036","target":"record","created_at":"2026-07-05T10:06:51Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"680fe66401d26ec029696164053ea150fa9c1a99b12cef8b79e8ddbaa5ed3859","cross_cats_sorted":["cs.CR","cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-01-28T21:52:15Z","title_canon_sha256":"8d07e8671a82fba664665b6085f921bb284b2ffc041757518114103345aa1629"},"schema_version":"1.0","source":{"id":"2501.17315","kind":"arxiv","version":1}},"canonical_sha256":"2e29b2ce30791dbc4c05f08cab160d9c051342406b05887b5a190465d6cb2a1a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2e29b2ce30791dbc4c05f08cab160d9c051342406b05887b5a190465d6cb2a1a","first_computed_at":"2026-07-05T10:06:51.161131Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:06:51.161131Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3CnW10hAY65Ln387ReHl8klaU7T2XdPbAo1f9bwJoUnyTnnxWs8S7PJUGFl84kARMn0derWZSD+VGIcZ5eJsAg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:06:51.161624Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.17315","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d79e9ca8720889329e41300668e9d1710b74dd57b9aaa7054a54761f83327036","sha256:737809d1ac6c7bffa423ae297f27a2e90821050c0aace9a59281588f358ba077"],"state_sha256":"a2fe3f3f1e3f6532c8ec132a6ada9eca95ea06a062cf1500bd187e648dcb1902"}