{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:UWLJ7VN5JNF4AEEXDKUDFBGEYL","short_pith_number":"pith:UWLJ7VN5","canonical_record":{"source":{"id":"2607.21735","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T18:34:12Z","cross_cats_sorted":["cs.CR"],"title_canon_sha256":"22d0ae5ff63be9458da3be145227c92bbcc1dd9014ca3ab272f58c2ac2cfacc4","abstract_canon_sha256":"a748418608ac6900ed2829adb3b1c1c51cfe807eca6fde1d2427cb825a1ed87e"},"schema_version":"1.0"},"canonical_sha256":"a5969fd5bd4b4bc010971aa83284c4c2d03f04ac696abb3b263a735ca00b6728","source":{"kind":"arxiv","id":"2607.21735","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.21735","created_at":"2026-07-27T00:20:18Z"},{"alias_kind":"arxiv_version","alias_value":"2607.21735v1","created_at":"2026-07-27T00:20:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.21735","created_at":"2026-07-27T00:20:18Z"},{"alias_kind":"pith_short_12","alias_value":"UWLJ7VN5JNF4","created_at":"2026-07-27T00:20:18Z"},{"alias_kind":"pith_short_16","alias_value":"UWLJ7VN5JNF4AEEX","created_at":"2026-07-27T00:20:18Z"},{"alias_kind":"pith_short_8","alias_value":"UWLJ7VN5","created_at":"2026-07-27T00:20:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:UWLJ7VN5JNF4AEEXDKUDFBGEYL","target":"record","payload":{"canonical_record":{"source":{"id":"2607.21735","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T18:34:12Z","cross_cats_sorted":["cs.CR"],"title_canon_sha256":"22d0ae5ff63be9458da3be145227c92bbcc1dd9014ca3ab272f58c2ac2cfacc4","abstract_canon_sha256":"a748418608ac6900ed2829adb3b1c1c51cfe807eca6fde1d2427cb825a1ed87e"},"schema_version":"1.0"},"canonical_sha256":"a5969fd5bd4b4bc010971aa83284c4c2d03f04ac696abb3b263a735ca00b6728","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-27T00:20:18.147907Z","signature_b64":"sP023UUwWdTlyPN7ub4od3V8XLoBsA6oucvGAcAph0vVn8Kmzth6ltZjyxMWkZuDpymvVieRhFaTSKlhDBoVDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a5969fd5bd4b4bc010971aa83284c4c2d03f04ac696abb3b263a735ca00b6728","last_reissued_at":"2026-07-27T00:20:18.147083Z","signature_status":"signed_v1","first_computed_at":"2026-07-27T00:20:18.147083Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.21735","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-27T00:20:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pRno30FCrGQ+PrzMg9qiAfXnuZbgqPjLpteOiywwTB7RXIh/+PGapaFTH5n7iXp8wTTu0z1a3dme5mAmXjzsCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T15:51:22.541724Z"},"content_sha256":"dccf374981cf81c7ff3ccf99cfd49fe74a11cd16f6f45b0c169bf79446cf95bb","schema_version":"1.0","event_id":"sha256:dccf374981cf81c7ff3ccf99cfd49fe74a11cd16f6f45b0c169bf79446cf95bb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:UWLJ7VN5JNF4AEEXDKUDFBGEYL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"What AI Red-Team Evaluations Can and Cannot Prove","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.AI","authors_text":"Bandana Kaur","submitted_at":"2026-07-23T18:34:12Z","abstract_excerpt":"Red-team evaluations of AI models support some claims and not others, and the boundary between the two is calculable rather than merely a matter of judgment. We define the evidential ceiling of an evaluation as the largest factor by which one result can move belief under a fixed testing budget, derive it in closed form for the benchmark null result, and use it to locate that boundary exactly. We find that above a calculable harm rate, a benchmark of modest size certifies a category to a stated evidentiary standard, and a clean sheet is then the stronger of the two possible observations, outwei"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.21735","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/2607.21735/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-27T00:20:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NOCYVr6Q4xhh1ls1GG9LJ8XMgdx6VR2q8IfJ+pUmR1bJ6HSqGZcySg6d36UW7m9sE+e7Nhy8GXWlscoHtt4xBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T15:51:22.542329Z"},"content_sha256":"51db08a6d3bb708128cc00a5fb2baaa130adb4a6a0ce32930a122d4021ee3dbb","schema_version":"1.0","event_id":"sha256:51db08a6d3bb708128cc00a5fb2baaa130adb4a6a0ce32930a122d4021ee3dbb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UWLJ7VN5JNF4AEEXDKUDFBGEYL/bundle.json","state_url":"https://pith.science/pith/UWLJ7VN5JNF4AEEXDKUDFBGEYL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UWLJ7VN5JNF4AEEXDKUDFBGEYL/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-06T15:51:22Z","links":{"resolver":"https://pith.science/pith/UWLJ7VN5JNF4AEEXDKUDFBGEYL","bundle":"https://pith.science/pith/UWLJ7VN5JNF4AEEXDKUDFBGEYL/bundle.json","state":"https://pith.science/pith/UWLJ7VN5JNF4AEEXDKUDFBGEYL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UWLJ7VN5JNF4AEEXDKUDFBGEYL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:UWLJ7VN5JNF4AEEXDKUDFBGEYL","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":"a748418608ac6900ed2829adb3b1c1c51cfe807eca6fde1d2427cb825a1ed87e","cross_cats_sorted":["cs.CR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T18:34:12Z","title_canon_sha256":"22d0ae5ff63be9458da3be145227c92bbcc1dd9014ca3ab272f58c2ac2cfacc4"},"schema_version":"1.0","source":{"id":"2607.21735","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.21735","created_at":"2026-07-27T00:20:18Z"},{"alias_kind":"arxiv_version","alias_value":"2607.21735v1","created_at":"2026-07-27T00:20:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.21735","created_at":"2026-07-27T00:20:18Z"},{"alias_kind":"pith_short_12","alias_value":"UWLJ7VN5JNF4","created_at":"2026-07-27T00:20:18Z"},{"alias_kind":"pith_short_16","alias_value":"UWLJ7VN5JNF4AEEX","created_at":"2026-07-27T00:20:18Z"},{"alias_kind":"pith_short_8","alias_value":"UWLJ7VN5","created_at":"2026-07-27T00:20:18Z"}],"graph_snapshots":[{"event_id":"sha256:51db08a6d3bb708128cc00a5fb2baaa130adb4a6a0ce32930a122d4021ee3dbb","target":"graph","created_at":"2026-07-27T00:20:18Z","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/2607.21735/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Red-team evaluations of AI models support some claims and not others, and the boundary between the two is calculable rather than merely a matter of judgment. We define the evidential ceiling of an evaluation as the largest factor by which one result can move belief under a fixed testing budget, derive it in closed form for the benchmark null result, and use it to locate that boundary exactly. We find that above a calculable harm rate, a benchmark of modest size certifies a category to a stated evidentiary standard, and a clean sheet is then the stronger of the two possible observations, outwei","authors_text":"Bandana Kaur","cross_cats":["cs.CR"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T18:34:12Z","title":"What AI Red-Team Evaluations Can and Cannot Prove"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.21735","kind":"arxiv","version":1},"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:dccf374981cf81c7ff3ccf99cfd49fe74a11cd16f6f45b0c169bf79446cf95bb","target":"record","created_at":"2026-07-27T00:20:18Z","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":"a748418608ac6900ed2829adb3b1c1c51cfe807eca6fde1d2427cb825a1ed87e","cross_cats_sorted":["cs.CR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T18:34:12Z","title_canon_sha256":"22d0ae5ff63be9458da3be145227c92bbcc1dd9014ca3ab272f58c2ac2cfacc4"},"schema_version":"1.0","source":{"id":"2607.21735","kind":"arxiv","version":1}},"canonical_sha256":"a5969fd5bd4b4bc010971aa83284c4c2d03f04ac696abb3b263a735ca00b6728","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a5969fd5bd4b4bc010971aa83284c4c2d03f04ac696abb3b263a735ca00b6728","first_computed_at":"2026-07-27T00:20:18.147083Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-27T00:20:18.147083Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"sP023UUwWdTlyPN7ub4od3V8XLoBsA6oucvGAcAph0vVn8Kmzth6ltZjyxMWkZuDpymvVieRhFaTSKlhDBoVDg==","signature_status":"signed_v1","signed_at":"2026-07-27T00:20:18.147907Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.21735","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dccf374981cf81c7ff3ccf99cfd49fe74a11cd16f6f45b0c169bf79446cf95bb","sha256:51db08a6d3bb708128cc00a5fb2baaa130adb4a6a0ce32930a122d4021ee3dbb"],"state_sha256":"9800048f79c8198aaa47d5718e106c94cf7983bf27e5af42112f96635d44a8c8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"t2g29XkAIzSOdT9ebYHgIls9kPGs07xlHoMNSC6x/9/TP8YhKcUJuzs+Rv2astz1BzmX45SLzl9QX5ZqbW0NAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T15:51:22.547656Z","bundle_sha256":"e5600112e5dcee2bd6a976cd543c0092b3631a7cb0f29b0b905aba84d9ac2af7"}}