{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:X27JNRHR7RM6Y4Q4XCNL5OGAZB","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":"8d9d7006bbc23da3d5e6884c806bee3bd9e34ab106c88ad868d14b9d483cc8b7","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-18T15:01:00Z","title_canon_sha256":"b66488cd07f7e564361cdae078901703f66e39b8b4e36c4bf9109bde16143f1a"},"schema_version":"1.0","source":{"id":"2404.12241","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.12241","created_at":"2026-07-05T08:18:51Z"},{"alias_kind":"arxiv_version","alias_value":"2404.12241v2","created_at":"2026-07-05T08:18:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.12241","created_at":"2026-07-05T08:18:51Z"},{"alias_kind":"pith_short_12","alias_value":"X27JNRHR7RM6","created_at":"2026-07-05T08:18:51Z"},{"alias_kind":"pith_short_16","alias_value":"X27JNRHR7RM6Y4Q4","created_at":"2026-07-05T08:18:51Z"},{"alias_kind":"pith_short_8","alias_value":"X27JNRHR","created_at":"2026-07-05T08:18:51Z"}],"graph_snapshots":[{"event_id":"sha256:7dfac2f7bb3f44fa3d588498fa114e68015bfa875ceb0a7ffc6aaf9b88506db0","target":"graph","created_at":"2026-07-05T08:18:51Z","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/2404.12241/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper introduces v0.5 of the AI Safety Benchmark, which has been created by the MLCommons AI Safety Working Group. The AI Safety Benchmark has been designed to assess the safety risks of AI systems that use chat-tuned language models. We introduce a principled approach to specifying and constructing the benchmark, which for v0.5 covers only a single use case (an adult chatting to a general-purpose assistant in English), and a limited set of personas (i.e., typical users, malicious users, and vulnerable users). We created a new taxonomy of 13 hazard categories, of which 7 have tests in the","authors_text":"Abolfazl Shahbazi, Adarsh Agrawal, Adina Williams, Agasthya Gangavarapu, Ahmed M. Ahmed, Alice Schoenauer Sebag, Alicia Parrish, Amin Nikanjam, Ananya Gangavarapu, Bertie Vidgen, Besmira Nushi, Bhavya Kailkhura, Bo Li, Borhane Blili-Hamelin, Brian Fuller, Canyu Chen, Carole-Jean Wu, Chris Knotz, Chris Lengerich, Chris Welty, Cigdem Patlak, Cody Coleman, Davide Testuggine, Debojyoti Dutta, Dinesh Jinenhally Naganna, Eileen Peters Long, Eleonora Presani, Elie Alhajjar, Elizabeth Anne Watkins, Emanuel Moss, Fabrizio Puletti, Fedor Zhdanov, Felix Juefei-Xu, Forough Poursabzi-Sangdeh, Foutse Khomh, Hannah Rose Kirk, Heather Frase, Ian Eisenberg, Iftach Orr, James Ezick, James Gealy, James Goel, Jiacheng Zhu, Joaquin Vanschoren, Joseph Marvin Imperial, Kal Chakra, Kelvin Manyeki, Kevin Klyman, Kurt Bollacker, Lama Nachman, Leonard Tang, Leon Derczynski, Lora Aroyo, Luis Oala, Marisa Ferrara Boston, Max Bartolo, Michael Kuchnik, Najla Alfaraj, Namir Al-Nuaimi, Nick Judd, Nino Scherrer, Patrick Schramowski, Paul R\\\"ottger, Percy Liang, Peter Mattson, Poonam Yadav, Priyanka Mary Mammen, Rajat Ghosh, Ram Gandikota, Rebecca Weiss, Rishi Bomassani, Sarah Luger, Saurav Sahay, Scott A. Hale, Sean McGregor, Shachi H. Kumar, Shafee Mohammed, Sim\\'eon Campos, Srijan Kumar, Sujata Goswami, Surgan Jandial, Tim Santos, Trupti Bavalatti, Tyler Wilbers, Usman Gohar, Vamsi Sistla, Victor Akinwande, Victor Lu, Vin Sharma, Virendra Mehta, Vithursan Thangarasa, Wenhui Zhang, Wiebke Hutiri, William Pietri, Xianjun Yang, Xudong Shen, Yifan Mai, Yi Zeng, Zacharie Delpierre Coudert, Zeyi Liao","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-18T15:01:00Z","title":"Introducing v0.5 of the AI Safety Benchmark from MLCommons"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.12241","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:d70a655b3f3c6b13b705fe6789b58dd853f9240ff2fe68c62715b606b593d368","target":"record","created_at":"2026-07-05T08:18:51Z","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":"8d9d7006bbc23da3d5e6884c806bee3bd9e34ab106c88ad868d14b9d483cc8b7","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-18T15:01:00Z","title_canon_sha256":"b66488cd07f7e564361cdae078901703f66e39b8b4e36c4bf9109bde16143f1a"},"schema_version":"1.0","source":{"id":"2404.12241","kind":"arxiv","version":2}},"canonical_sha256":"bebe96c4f1fc59ec721cb89abeb8c0c87a76316b3ff685ab98089ef9b2325f9e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bebe96c4f1fc59ec721cb89abeb8c0c87a76316b3ff685ab98089ef9b2325f9e","first_computed_at":"2026-07-05T08:18:51.887654Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:18:51.887654Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wj97qEzc+RdePA8Lhojm+xW1Mvs5CQc+TG2T2QQM9T0wQTLmGAFy0JKuM1IC8cPygJyu9CeK4rutCaPMM5MxBA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:18:51.888138Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.12241","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d70a655b3f3c6b13b705fe6789b58dd853f9240ff2fe68c62715b606b593d368","sha256:7dfac2f7bb3f44fa3d588498fa114e68015bfa875ceb0a7ffc6aaf9b88506db0"],"state_sha256":"fbb2560d3313f8bb3da15e35b1383733e45b6e0977225b59ed06de389a90e64e"}