{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:J5OZ2G562AXEITCD7ZTP2KXSAY","short_pith_number":"pith:J5OZ2G56","schema_version":"1.0","canonical_sha256":"4f5d9d1bbed02e444c43fe66fd2af206136f5322890b846124848350832f71ca","source":{"kind":"arxiv","id":"2607.22671","version":1},"attestation_state":"computed","paper":{"title":"AIR-BENCH Live: An Evolving Safety Benchmark for Foundation Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bo Li, Minzhou Pan, Rohan Naphade","submitted_at":"2026-07-06T17:26:46Z","abstract_excerpt":"Foundation-model safety benchmarks capture the AI risks of their time of publication: as models improve and governments pass new AI-safety legislation, their risk taxonomies become incomprehensive and their attack prompts become ineffective. We present AIR-BENCH Live, a self-evolving successor to AIR-BENCH 2024. An automated update pipeline monitors government regulation and classifies new policies against the current four-tier risk taxonomy, either matching them to existing categories or proposing new granular categories. Then, a multi-agent, persona-driven prompt generation algorithm generat"},"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":"2607.22671","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-06T17:26:46Z","cross_cats_sorted":[],"title_canon_sha256":"af317ba149a8710fcd324c91665da20d71a9438ce419f60a52ac0aa862df4f8e","abstract_canon_sha256":"bdb23cf5b7e0f49c02f7d8c9ab6155160db6ded25eff24b8e4b3f3740da0490a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T00:21:48.325119Z","signature_b64":"K87V2z2FN74fLnxzst+RLd05OhLwQUKRZ2mYhuvBunUl/s7KY+41Jlfk/lNVmdhu0O6/02n1LXq46rSfuZGTBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4f5d9d1bbed02e444c43fe66fd2af206136f5322890b846124848350832f71ca","last_reissued_at":"2026-07-28T00:21:48.324165Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T00:21:48.324165Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AIR-BENCH Live: An Evolving Safety Benchmark for Foundation Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bo Li, Minzhou Pan, Rohan Naphade","submitted_at":"2026-07-06T17:26:46Z","abstract_excerpt":"Foundation-model safety benchmarks capture the AI risks of their time of publication: as models improve and governments pass new AI-safety legislation, their risk taxonomies become incomprehensive and their attack prompts become ineffective. We present AIR-BENCH Live, a self-evolving successor to AIR-BENCH 2024. An automated update pipeline monitors government regulation and classifies new policies against the current four-tier risk taxonomy, either matching them to existing categories or proposing new granular categories. Then, a multi-agent, persona-driven prompt generation algorithm generat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.22671","kind":"arxiv","version":1},"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/2607.22671/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":"2607.22671","created_at":"2026-07-28T00:21:48.324621+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.22671v1","created_at":"2026-07-28T00:21:48.324621+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.22671","created_at":"2026-07-28T00:21:48.324621+00:00"},{"alias_kind":"pith_short_12","alias_value":"J5OZ2G562AXE","created_at":"2026-07-28T00:21:48.324621+00:00"},{"alias_kind":"pith_short_16","alias_value":"J5OZ2G562AXEITCD","created_at":"2026-07-28T00:21:48.324621+00:00"},{"alias_kind":"pith_short_8","alias_value":"J5OZ2G56","created_at":"2026-07-28T00:21:48.324621+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/J5OZ2G562AXEITCD7ZTP2KXSAY","json":"https://pith.science/pith/J5OZ2G562AXEITCD7ZTP2KXSAY.json","graph_json":"https://pith.science/api/pith-number/J5OZ2G562AXEITCD7ZTP2KXSAY/graph.json","events_json":"https://pith.science/api/pith-number/J5OZ2G562AXEITCD7ZTP2KXSAY/events.json","paper":"https://pith.science/paper/J5OZ2G56"},"agent_actions":{"view_html":"https://pith.science/pith/J5OZ2G562AXEITCD7ZTP2KXSAY","download_json":"https://pith.science/pith/J5OZ2G562AXEITCD7ZTP2KXSAY.json","view_paper":"https://pith.science/paper/J5OZ2G56","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.22671&json=true","fetch_graph":"https://pith.science/api/pith-number/J5OZ2G562AXEITCD7ZTP2KXSAY/graph.json","fetch_events":"https://pith.science/api/pith-number/J5OZ2G562AXEITCD7ZTP2KXSAY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/J5OZ2G562AXEITCD7ZTP2KXSAY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/J5OZ2G562AXEITCD7ZTP2KXSAY/action/storage_attestation","attest_author":"https://pith.science/pith/J5OZ2G562AXEITCD7ZTP2KXSAY/action/author_attestation","sign_citation":"https://pith.science/pith/J5OZ2G562AXEITCD7ZTP2KXSAY/action/citation_signature","submit_replication":"https://pith.science/pith/J5OZ2G562AXEITCD7ZTP2KXSAY/action/replication_record"}},"created_at":"2026-07-28T00:21:48.324621+00:00","updated_at":"2026-07-28T00:21:48.324621+00:00"}