{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:76SF6W63VX6E7XYC3SXNZ2NCGK","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":"19eb0474fc831581c1e3e575b9c184e0cd6a0bc7665a2a4d690bd5165dbb1013","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-06-20T17:43:18Z","title_canon_sha256":"aaaf8bb894ce9ee6d2eebbcb981095d48a271d16cec89eac401028ab570eb79a"},"schema_version":"1.0","source":{"id":"2406.14595","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.14595","created_at":"2026-07-05T08:39:00Z"},{"alias_kind":"arxiv_version","alias_value":"2406.14595v2","created_at":"2026-07-05T08:39:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.14595","created_at":"2026-07-05T08:39:00Z"},{"alias_kind":"pith_short_12","alias_value":"76SF6W63VX6E","created_at":"2026-07-05T08:39:00Z"},{"alias_kind":"pith_short_16","alias_value":"76SF6W63VX6E7XYC","created_at":"2026-07-05T08:39:00Z"},{"alias_kind":"pith_short_8","alias_value":"76SF6W63","created_at":"2026-07-05T08:39:00Z"}],"graph_snapshots":[{"event_id":"sha256:460a646e185784f2c8f9e0f9e1f931da086b4d1772ba0f7fe341713b2b7d9fd0","target":"graph","created_at":"2026-07-05T08:39:00Z","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/2406.14595/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Developers try to evaluate whether an AI system can be misused by adversaries before releasing it; for example, they might test whether a model enables cyberoffense, user manipulation, or bioterrorism. In this work, we show that individually testing models for misuse is inadequate; adversaries can misuse combinations of models even when each individual model is safe. The adversary accomplishes this by first decomposing tasks into subtasks, then solving each subtask with the best-suited model. For example, an adversary might solve challenging-but-benign subtasks with an aligned frontier model, ","authors_text":"Anca Dragan, Erik Jones, Jacob Steinhardt","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-06-20T17:43:18Z","title":"Adversaries Can Misuse Combinations of Safe Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.14595","kind":"arxiv","version":2},"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":"sha256:64059844bc21d44da2369ba1be3a0fe582438635e77a8da8af19a720127aef77","target":"record","created_at":"2026-07-05T08:39:00Z","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":"19eb0474fc831581c1e3e575b9c184e0cd6a0bc7665a2a4d690bd5165dbb1013","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-06-20T17:43:18Z","title_canon_sha256":"aaaf8bb894ce9ee6d2eebbcb981095d48a271d16cec89eac401028ab570eb79a"},"schema_version":"1.0","source":{"id":"2406.14595","kind":"arxiv","version":2}},"canonical_sha256":"ffa45f5bdbadfc4fdf02dcaedce9a232a5a6a1b4c55ce64bb482c7c6376ee71d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ffa45f5bdbadfc4fdf02dcaedce9a232a5a6a1b4c55ce64bb482c7c6376ee71d","first_computed_at":"2026-07-05T08:39:00.679827Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:39:00.679827Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Sl9+5vRlfVhRnvQLC/D9W0s5FSQHbFNqjZjrli0cdP3wra63R+RPTZoAKx5VGvsOIpMenoQX+gfJrIHXL5aBAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:39:00.680346Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.14595","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:64059844bc21d44da2369ba1be3a0fe582438635e77a8da8af19a720127aef77","sha256:460a646e185784f2c8f9e0f9e1f931da086b4d1772ba0f7fe341713b2b7d9fd0"],"state_sha256":"1f30b4dc6cab22347e6062c5bffbd1ca67a83bbc7eedfdfd92bd4da292350eb2"}