{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:7C3XCHEA62W42YRUCIGAJCDUQS","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":"8ca65090493239672170dc4deeb9ed71d06cc296d80d1234463460a6b60000a6","cross_cats_sorted":["cs.GT","cs.LG","cs.MA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-12-23T18:53:33Z","title_canon_sha256":"56535b9e6b81018f82d6b5975d3babeb1b73d9996322760a8dd870efbf850645"},"schema_version":"1.0","source":{"id":"2412.17797","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.17797","created_at":"2026-07-05T11:50:58Z"},{"alias_kind":"arxiv_version","alias_value":"2412.17797v2","created_at":"2026-07-05T11:50:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.17797","created_at":"2026-07-05T11:50:58Z"},{"alias_kind":"pith_short_12","alias_value":"7C3XCHEA62W4","created_at":"2026-07-05T11:50:58Z"},{"alias_kind":"pith_short_16","alias_value":"7C3XCHEA62W42YRU","created_at":"2026-07-05T11:50:58Z"},{"alias_kind":"pith_short_8","alias_value":"7C3XCHEA","created_at":"2026-07-05T11:50:58Z"}],"graph_snapshots":[{"event_id":"sha256:38e547d608261bbe7ec1f74332cf0bc7583faa3774be45272249f9f372075fd6","target":"graph","created_at":"2026-07-05T11:50:58Z","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/2412.17797/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We study partially observable assistance games (POAGs), a model of the human-AI value alignment problem which allows the human and the AI assistant to have partial observations. Motivated by concerns of AI deception, we study a qualitatively new phenomenon made possible by partial observability: would an AI assistant ever have an incentive to interfere with the human's observations? First, we prove that sometimes an optimal assistant must take observation-interfering actions, even when the human is playing optimally, and even when there are otherwise-equivalent actions available that do not in","authors_text":"Caspar Oesterheld, Scott Emmons, Stuart Russell, Vincent Conitzer","cross_cats":["cs.GT","cs.LG","cs.MA"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-12-23T18:53:33Z","title":"Observation Interference in Partially Observable Assistance Games"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.17797","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":"sha256:0d4f01ed1f436c71071e71ed6ef539446e75ff7eb331b1a19a6da1cab2908c35","target":"record","created_at":"2026-07-05T11:50:58Z","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":"8ca65090493239672170dc4deeb9ed71d06cc296d80d1234463460a6b60000a6","cross_cats_sorted":["cs.GT","cs.LG","cs.MA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-12-23T18:53:33Z","title_canon_sha256":"56535b9e6b81018f82d6b5975d3babeb1b73d9996322760a8dd870efbf850645"},"schema_version":"1.0","source":{"id":"2412.17797","kind":"arxiv","version":2}},"canonical_sha256":"f8b7711c80f6adcd6234120c04887484ae5b9950050aadc769692d8f39124ed4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f8b7711c80f6adcd6234120c04887484ae5b9950050aadc769692d8f39124ed4","first_computed_at":"2026-07-05T11:50:58.690262Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:50:58.690262Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"iT9Y3Cn0EM5zNbRBt+f5XjVAo6+jk5H46+UgyXQP2UaTyxriX5GBxeURmDfZ97QNKaRvnVvxux4Y0VPETWC2Dg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:50:58.690766Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.17797","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0d4f01ed1f436c71071e71ed6ef539446e75ff7eb331b1a19a6da1cab2908c35","sha256:38e547d608261bbe7ec1f74332cf0bc7583faa3774be45272249f9f372075fd6"],"state_sha256":"0c9dbecaf96119592cef1e3982a65c9cbf08dd6ea8c07a2524cd30ec72499e48"}