{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:KJ6L7M4435LDI5B6GBN65BTXGU","short_pith_number":"pith:KJ6L7M44","canonical_record":{"source":{"id":"2312.01350","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-12-03T11:11:57Z","cross_cats_sorted":[],"title_canon_sha256":"a95b9365ff70eea0ef98eb2d71908477cd1bc0dee23ec30be5816c4705c8a1a0","abstract_canon_sha256":"dbae5f8f8d1eabce4f98745f533684a96b1af87dff98e627aeb5320db2926b38"},"schema_version":"1.0"},"canonical_sha256":"527cbfb39cdf5634743e305bee867735087d17f10c71c66e2e0f2b70a676ba8f","source":{"kind":"arxiv","id":"2312.01350","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.01350","created_at":"2026-07-05T07:19:33Z"},{"alias_kind":"arxiv_version","alias_value":"2312.01350v1","created_at":"2026-07-05T07:19:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.01350","created_at":"2026-07-05T07:19:33Z"},{"alias_kind":"pith_short_12","alias_value":"KJ6L7M4435LD","created_at":"2026-07-05T07:19:33Z"},{"alias_kind":"pith_short_16","alias_value":"KJ6L7M4435LDI5B6","created_at":"2026-07-05T07:19:33Z"},{"alias_kind":"pith_short_8","alias_value":"KJ6L7M44","created_at":"2026-07-05T07:19:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:KJ6L7M4435LDI5B6GBN65BTXGU","target":"record","payload":{"canonical_record":{"source":{"id":"2312.01350","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-12-03T11:11:57Z","cross_cats_sorted":[],"title_canon_sha256":"a95b9365ff70eea0ef98eb2d71908477cd1bc0dee23ec30be5816c4705c8a1a0","abstract_canon_sha256":"dbae5f8f8d1eabce4f98745f533684a96b1af87dff98e627aeb5320db2926b38"},"schema_version":"1.0"},"canonical_sha256":"527cbfb39cdf5634743e305bee867735087d17f10c71c66e2e0f2b70a676ba8f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:19:33.683225Z","signature_b64":"XI+mQvacDYJ8W/WReNItgjpFeRNNgaOCnIzerkLkVdM/zDhPYlsRHuBq/QdbW4gaXryn7/ZgzJsesI+F27ZzCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"527cbfb39cdf5634743e305bee867735087d17f10c71c66e2e0f2b70a676ba8f","last_reissued_at":"2026-07-05T07:19:33.682722Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:19:33.682722Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2312.01350","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":"2026-07-05T07:19:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Sf3qQ73rCk4NPohUQSmDs1lMqkmfhkSF0eWA4R7fR/SI4o2XhHzQ/oi/zXQSc9IkpzUmSoMRGSlN5hLvSoXoBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T22:22:12.452052Z"},"content_sha256":"57f73881dde17641162f38c8b09b8286556d7e12f4f959b6d2ea2afe463f1b66","schema_version":"1.0","event_id":"sha256:57f73881dde17641162f38c8b09b8286556d7e12f4f959b6d2ea2afe463f1b66"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:KJ6L7M4435LDI5B6GBN65BTXGU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Honesty Is the Best Policy: Defining and Mitigating AI Deception","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Francesca Toni, Francesco Belardinelli, Francis Rhys Ward, Tom Everitt","submitted_at":"2023-12-03T11:11:57Z","abstract_excerpt":"Deceptive agents are a challenge for the safety, trustworthiness, and cooperation of AI systems. We focus on the problem that agents might deceive in order to achieve their goals (for instance, in our experiments with language models, the goal of being evaluated as truthful). There are a number of existing definitions of deception in the literature on game theory and symbolic AI, but there is no overarching theory of deception for learning agents in games. We introduce a formal definition of deception in structural causal games, grounded in the philosophy literature, and applicable to real-wor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.01350","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/2312.01350/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":"2026-07-05T07:19:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XZLPE2DY2voce+ijKyu5e9bAZuTU1U90QoWe9noE5NwcuN1ATCern6NdR+Jrl8PNePhGaL99rdmv7KBEN0q2BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T22:22:12.453048Z"},"content_sha256":"09266459908890d16d50bb3355c94433536fc704a6c6c2625604d23d888e48bc","schema_version":"1.0","event_id":"sha256:09266459908890d16d50bb3355c94433536fc704a6c6c2625604d23d888e48bc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KJ6L7M4435LDI5B6GBN65BTXGU/bundle.json","state_url":"https://pith.science/pith/KJ6L7M4435LDI5B6GBN65BTXGU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KJ6L7M4435LDI5B6GBN65BTXGU/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":"2026-08-15T22:22:12Z","links":{"resolver":"https://pith.science/pith/KJ6L7M4435LDI5B6GBN65BTXGU","bundle":"https://pith.science/pith/KJ6L7M4435LDI5B6GBN65BTXGU/bundle.json","state":"https://pith.science/pith/KJ6L7M4435LDI5B6GBN65BTXGU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KJ6L7M4435LDI5B6GBN65BTXGU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:KJ6L7M4435LDI5B6GBN65BTXGU","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":"dbae5f8f8d1eabce4f98745f533684a96b1af87dff98e627aeb5320db2926b38","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-12-03T11:11:57Z","title_canon_sha256":"a95b9365ff70eea0ef98eb2d71908477cd1bc0dee23ec30be5816c4705c8a1a0"},"schema_version":"1.0","source":{"id":"2312.01350","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.01350","created_at":"2026-07-05T07:19:33Z"},{"alias_kind":"arxiv_version","alias_value":"2312.01350v1","created_at":"2026-07-05T07:19:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.01350","created_at":"2026-07-05T07:19:33Z"},{"alias_kind":"pith_short_12","alias_value":"KJ6L7M4435LD","created_at":"2026-07-05T07:19:33Z"},{"alias_kind":"pith_short_16","alias_value":"KJ6L7M4435LDI5B6","created_at":"2026-07-05T07:19:33Z"},{"alias_kind":"pith_short_8","alias_value":"KJ6L7M44","created_at":"2026-07-05T07:19:33Z"}],"graph_snapshots":[{"event_id":"sha256:09266459908890d16d50bb3355c94433536fc704a6c6c2625604d23d888e48bc","target":"graph","created_at":"2026-07-05T07:19:33Z","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/2312.01350/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Deceptive agents are a challenge for the safety, trustworthiness, and cooperation of AI systems. We focus on the problem that agents might deceive in order to achieve their goals (for instance, in our experiments with language models, the goal of being evaluated as truthful). There are a number of existing definitions of deception in the literature on game theory and symbolic AI, but there is no overarching theory of deception for learning agents in games. We introduce a formal definition of deception in structural causal games, grounded in the philosophy literature, and applicable to real-wor","authors_text":"Francesca Toni, Francesco Belardinelli, Francis Rhys Ward, Tom Everitt","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-12-03T11:11:57Z","title":"Honesty Is the Best Policy: Defining and Mitigating AI Deception"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.01350","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":"sha256:57f73881dde17641162f38c8b09b8286556d7e12f4f959b6d2ea2afe463f1b66","target":"record","created_at":"2026-07-05T07:19:33Z","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":"dbae5f8f8d1eabce4f98745f533684a96b1af87dff98e627aeb5320db2926b38","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-12-03T11:11:57Z","title_canon_sha256":"a95b9365ff70eea0ef98eb2d71908477cd1bc0dee23ec30be5816c4705c8a1a0"},"schema_version":"1.0","source":{"id":"2312.01350","kind":"arxiv","version":1}},"canonical_sha256":"527cbfb39cdf5634743e305bee867735087d17f10c71c66e2e0f2b70a676ba8f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"527cbfb39cdf5634743e305bee867735087d17f10c71c66e2e0f2b70a676ba8f","first_computed_at":"2026-07-05T07:19:33.682722Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:19:33.682722Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XI+mQvacDYJ8W/WReNItgjpFeRNNgaOCnIzerkLkVdM/zDhPYlsRHuBq/QdbW4gaXryn7/ZgzJsesI+F27ZzCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:19:33.683225Z","signed_message":"canonical_sha256_bytes"},"source_id":"2312.01350","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:57f73881dde17641162f38c8b09b8286556d7e12f4f959b6d2ea2afe463f1b66","sha256:09266459908890d16d50bb3355c94433536fc704a6c6c2625604d23d888e48bc"],"state_sha256":"915e422c7d1da0cf20fa706ae736e5f89f6d049e14bcca54cb86d517a4cc1e92"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"m/uzkEnFKTeWE4MTkXTbjkIzw6iI8IhOu5lc155xtEsZz0FXjOCQ2E2i7NjjaPI2LwE8RyMCcEP+t5+5HpahDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-15T22:22:12.458956Z","bundle_sha256":"0980ad0a013774525d1e6daca8f46706ca6085dcb39a66c02f60547a60f0be1e"}}