{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:OOL2AAXTOGBJ343AIT2CO5KG5Z","short_pith_number":"pith:OOL2AAXT","schema_version":"1.0","canonical_sha256":"7397a002f371829df36044f4277546ee4ee47b871fd77a104d7de8e83a77ef4c","source":{"kind":"arxiv","id":"2305.14965","version":4},"attestation_state":"computed","paper":{"title":"Tricking LLMs into Disobedience: Formalizing, Analyzing, and Detecting Jailbreaks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Abhinav Rao, Atharva Naik, Monojit Choudhury, Sachin Vashistha, Somak Aditya","submitted_at":"2023-05-24T09:57:37Z","abstract_excerpt":"Recent explorations with commercial Large Language Models (LLMs) have shown that non-expert users can jailbreak LLMs by simply manipulating their prompts; resulting in degenerate output behavior, privacy and security breaches, offensive outputs, and violations of content regulator policies. Limited studies have been conducted to formalize and analyze these attacks and their mitigations. We bridge this gap by proposing a formalism and a taxonomy of known (and possible) jailbreaks. We survey existing jailbreak methods and their effectiveness on open-source and commercial LLMs (such as GPT-based "},"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":"2305.14965","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-24T09:57:37Z","cross_cats_sorted":[],"title_canon_sha256":"4999a96535b1ced05dde3495dd692f47ce01dd7c75720bf1ce8fd4dec63df9e5","abstract_canon_sha256":"eb9a46eec941630a4da7347c207f65cfc4664bc403145f49e6f59b4881d9fd34"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:00:59.390884Z","signature_b64":"3QDxrjjN/1/AsD00ZBXyPGPrzjSzvZXpwbDoXwJh4iuHENKcAOrcM/Wz8xo2myiIFOR15hZZvMk5mBt8900AAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7397a002f371829df36044f4277546ee4ee47b871fd77a104d7de8e83a77ef4c","last_reissued_at":"2026-07-05T08:00:59.390404Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:00:59.390404Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Tricking LLMs into Disobedience: Formalizing, Analyzing, and Detecting Jailbreaks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Abhinav Rao, Atharva Naik, Monojit Choudhury, Sachin Vashistha, Somak Aditya","submitted_at":"2023-05-24T09:57:37Z","abstract_excerpt":"Recent explorations with commercial Large Language Models (LLMs) have shown that non-expert users can jailbreak LLMs by simply manipulating their prompts; resulting in degenerate output behavior, privacy and security breaches, offensive outputs, and violations of content regulator policies. Limited studies have been conducted to formalize and analyze these attacks and their mitigations. We bridge this gap by proposing a formalism and a taxonomy of known (and possible) jailbreaks. We survey existing jailbreak methods and their effectiveness on open-source and commercial LLMs (such as GPT-based "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14965","kind":"arxiv","version":4},"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/2305.14965/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":"2305.14965","created_at":"2026-07-05T08:00:59.390464+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.14965v4","created_at":"2026-07-05T08:00:59.390464+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14965","created_at":"2026-07-05T08:00:59.390464+00:00"},{"alias_kind":"pith_short_12","alias_value":"OOL2AAXTOGBJ","created_at":"2026-07-05T08:00:59.390464+00:00"},{"alias_kind":"pith_short_16","alias_value":"OOL2AAXTOGBJ343A","created_at":"2026-07-05T08:00:59.390464+00:00"},{"alias_kind":"pith_short_8","alias_value":"OOL2AAXT","created_at":"2026-07-05T08:00:59.390464+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.15425","citing_title":"Runtime-Structured Task Decomposition for Agentic Coding Systems","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2401.05561","citing_title":"TrustLLM: Trustworthiness in Large Language Models","ref_index":228,"is_internal_anchor":false},{"citing_arxiv_id":"2407.04295","citing_title":"Jailbreak Attacks and Defenses Against Large Language Models: A Survey","ref_index":72,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OOL2AAXTOGBJ343AIT2CO5KG5Z","json":"https://pith.science/pith/OOL2AAXTOGBJ343AIT2CO5KG5Z.json","graph_json":"https://pith.science/api/pith-number/OOL2AAXTOGBJ343AIT2CO5KG5Z/graph.json","events_json":"https://pith.science/api/pith-number/OOL2AAXTOGBJ343AIT2CO5KG5Z/events.json","paper":"https://pith.science/paper/OOL2AAXT"},"agent_actions":{"view_html":"https://pith.science/pith/OOL2AAXTOGBJ343AIT2CO5KG5Z","download_json":"https://pith.science/pith/OOL2AAXTOGBJ343AIT2CO5KG5Z.json","view_paper":"https://pith.science/paper/OOL2AAXT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.14965&json=true","fetch_graph":"https://pith.science/api/pith-number/OOL2AAXTOGBJ343AIT2CO5KG5Z/graph.json","fetch_events":"https://pith.science/api/pith-number/OOL2AAXTOGBJ343AIT2CO5KG5Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OOL2AAXTOGBJ343AIT2CO5KG5Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OOL2AAXTOGBJ343AIT2CO5KG5Z/action/storage_attestation","attest_author":"https://pith.science/pith/OOL2AAXTOGBJ343AIT2CO5KG5Z/action/author_attestation","sign_citation":"https://pith.science/pith/OOL2AAXTOGBJ343AIT2CO5KG5Z/action/citation_signature","submit_replication":"https://pith.science/pith/OOL2AAXTOGBJ343AIT2CO5KG5Z/action/replication_record"}},"created_at":"2026-07-05T08:00:59.390464+00:00","updated_at":"2026-07-05T08:00:59.390464+00:00"}