{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ZBAYK4UY5GWAGMLLSBZE7JA4DT","short_pith_number":"pith:ZBAYK4UY","schema_version":"1.0","canonical_sha256":"c841857298e9ac03316b90724fa41c1cfad54fca26e77544d4a9920724e38047","source":{"kind":"arxiv","id":"2406.06302","version":2},"attestation_state":"computed","paper":{"title":"Unveiling the Safety of GPT-4o: An Empirical Study using Jailbreak Attacks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CR","authors_text":"Aishan Liu, Dacheng Tao, Xianglong Liu, Zonghao Ying","submitted_at":"2024-06-10T14:18:56Z","abstract_excerpt":"The recent release of GPT-4o has garnered widespread attention due to its powerful general capabilities. While its impressive performance is widely acknowledged, its safety aspects have not been sufficiently explored. Given the potential societal impact of risky content generated by advanced generative AI such as GPT-4o, it is crucial to rigorously evaluate its safety. In response to this question, this paper for the first time conducts a rigorous evaluation of GPT-4o against jailbreak attacks. Specifically, this paper adopts a series of multi-modal and uni-modal jailbreak attacks on 4 commonl"},"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":"2406.06302","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-06-10T14:18:56Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"99bf8839c58c33dfb54b6601432e9a33067b97173a3633393ab0fd2a24d6e7d3","abstract_canon_sha256":"d1e683e0e07a5fea0ccebdca74f603cbe9481dfd7674d9d125aa399379186000"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:39:34.884547Z","signature_b64":"+LLPf4l1I5N9pi5665QS+iTkmyF0aH85EEPWctUCbf/NqjDyis5oMIfpoOb8YuLo7Ot20nS5T31bPXC60P27BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c841857298e9ac03316b90724fa41c1cfad54fca26e77544d4a9920724e38047","last_reissued_at":"2026-07-05T08:39:34.884054Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:39:34.884054Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unveiling the Safety of GPT-4o: An Empirical Study using Jailbreak Attacks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CR","authors_text":"Aishan Liu, Dacheng Tao, Xianglong Liu, Zonghao Ying","submitted_at":"2024-06-10T14:18:56Z","abstract_excerpt":"The recent release of GPT-4o has garnered widespread attention due to its powerful general capabilities. While its impressive performance is widely acknowledged, its safety aspects have not been sufficiently explored. Given the potential societal impact of risky content generated by advanced generative AI such as GPT-4o, it is crucial to rigorously evaluate its safety. In response to this question, this paper for the first time conducts a rigorous evaluation of GPT-4o against jailbreak attacks. Specifically, this paper adopts a series of multi-modal and uni-modal jailbreak attacks on 4 commonl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.06302","kind":"arxiv","version":2},"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/2406.06302/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":"2406.06302","created_at":"2026-07-05T08:39:34.884115+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.06302v2","created_at":"2026-07-05T08:39:34.884115+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.06302","created_at":"2026-07-05T08:39:34.884115+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZBAYK4UY5GWA","created_at":"2026-07-05T08:39:34.884115+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZBAYK4UY5GWAGMLL","created_at":"2026-07-05T08:39:34.884115+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZBAYK4UY","created_at":"2026-07-05T08:39:34.884115+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2507.21540","citing_title":"PRISM: Programmatic Reasoning with Image Sequence Manipulation for LVLM Jailbreaking","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2510.10073","citing_title":"SecureWebArena: A Holistic Security Evaluation Benchmark for LVLM-based Web Agents","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14604","citing_title":"Hijacking Large Audio-Language Models via Context-Agnostic and Imperceptible Auditory Prompt Injection","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZBAYK4UY5GWAGMLLSBZE7JA4DT","json":"https://pith.science/pith/ZBAYK4UY5GWAGMLLSBZE7JA4DT.json","graph_json":"https://pith.science/api/pith-number/ZBAYK4UY5GWAGMLLSBZE7JA4DT/graph.json","events_json":"https://pith.science/api/pith-number/ZBAYK4UY5GWAGMLLSBZE7JA4DT/events.json","paper":"https://pith.science/paper/ZBAYK4UY"},"agent_actions":{"view_html":"https://pith.science/pith/ZBAYK4UY5GWAGMLLSBZE7JA4DT","download_json":"https://pith.science/pith/ZBAYK4UY5GWAGMLLSBZE7JA4DT.json","view_paper":"https://pith.science/paper/ZBAYK4UY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.06302&json=true","fetch_graph":"https://pith.science/api/pith-number/ZBAYK4UY5GWAGMLLSBZE7JA4DT/graph.json","fetch_events":"https://pith.science/api/pith-number/ZBAYK4UY5GWAGMLLSBZE7JA4DT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZBAYK4UY5GWAGMLLSBZE7JA4DT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZBAYK4UY5GWAGMLLSBZE7JA4DT/action/storage_attestation","attest_author":"https://pith.science/pith/ZBAYK4UY5GWAGMLLSBZE7JA4DT/action/author_attestation","sign_citation":"https://pith.science/pith/ZBAYK4UY5GWAGMLLSBZE7JA4DT/action/citation_signature","submit_replication":"https://pith.science/pith/ZBAYK4UY5GWAGMLLSBZE7JA4DT/action/replication_record"}},"created_at":"2026-07-05T08:39:34.884115+00:00","updated_at":"2026-07-05T08:39:34.884115+00:00"}