{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:BE53Y3SPS4PO2EM2OBDAQZ3ZVI","short_pith_number":"pith:BE53Y3SP","schema_version":"1.0","canonical_sha256":"093bbc6e4f971eed119a7046086779aa365186aeeb858187859260b5d7f69f10","source":{"kind":"arxiv","id":"1908.01695","version":2},"attestation_state":"computed","paper":{"title":"Corrigibility with Utility Preservation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.MA"],"primary_cat":"cs.AI","authors_text":"Koen Holtman","submitted_at":"2019-08-05T15:40:45Z","abstract_excerpt":"Corrigibility is a safety property for artificially intelligent agents. A corrigible agent will not resist attempts by authorized parties to alter the goals and constraints that were encoded in the agent when it was first started. This paper shows how to construct a safety layer that adds corrigibility to arbitrarily advanced utility maximizing agents, including possible future agents with Artificial General Intelligence (AGI). The layer counter-acts the emergent incentive of advanced agents to resist such alteration. A detailed model for agents which can reason about preserving their utility "},"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":"1908.01695","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2019-08-05T15:40:45Z","cross_cats_sorted":["cs.MA"],"title_canon_sha256":"0040e1b8cdf6f01e3ab8f535f55f7cf37c398fbfa8f3974b202e763ecd4e01ba","abstract_canon_sha256":"c7b8ba6fb418382b3361005cbc9bf63f6da3e8c21e0dd2b67291a8894e65c101"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:52:27.166068Z","signature_b64":"rjGg1K94l3gexrpZpy0c/dQdE6Sr7qcsz+EvkxoxGphbc2Mg+vo4gaLShniCKNxNG8ODR3uhDmb67ClqHZVNAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"093bbc6e4f971eed119a7046086779aa365186aeeb858187859260b5d7f69f10","last_reissued_at":"2026-07-05T00:52:27.165635Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:52:27.165635Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Corrigibility with Utility Preservation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.MA"],"primary_cat":"cs.AI","authors_text":"Koen Holtman","submitted_at":"2019-08-05T15:40:45Z","abstract_excerpt":"Corrigibility is a safety property for artificially intelligent agents. A corrigible agent will not resist attempts by authorized parties to alter the goals and constraints that were encoded in the agent when it was first started. This paper shows how to construct a safety layer that adds corrigibility to arbitrarily advanced utility maximizing agents, including possible future agents with Artificial General Intelligence (AGI). The layer counter-acts the emergent incentive of advanced agents to resist such alteration. A detailed model for agents which can reason about preserving their utility "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1908.01695","kind":"arxiv","version":2},"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/1908.01695/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":"1908.01695","created_at":"2026-07-05T00:52:27.165696+00:00"},{"alias_kind":"arxiv_version","alias_value":"1908.01695v2","created_at":"2026-07-05T00:52:27.165696+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1908.01695","created_at":"2026-07-05T00:52:27.165696+00:00"},{"alias_kind":"pith_short_12","alias_value":"BE53Y3SPS4PO","created_at":"2026-07-05T00:52:27.165696+00:00"},{"alias_kind":"pith_short_16","alias_value":"BE53Y3SPS4PO2EM2","created_at":"2026-07-05T00:52:27.165696+00:00"},{"alias_kind":"pith_short_8","alias_value":"BE53Y3SP","created_at":"2026-07-05T00:52:27.165696+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/BE53Y3SPS4PO2EM2OBDAQZ3ZVI","json":"https://pith.science/pith/BE53Y3SPS4PO2EM2OBDAQZ3ZVI.json","graph_json":"https://pith.science/api/pith-number/BE53Y3SPS4PO2EM2OBDAQZ3ZVI/graph.json","events_json":"https://pith.science/api/pith-number/BE53Y3SPS4PO2EM2OBDAQZ3ZVI/events.json","paper":"https://pith.science/paper/BE53Y3SP"},"agent_actions":{"view_html":"https://pith.science/pith/BE53Y3SPS4PO2EM2OBDAQZ3ZVI","download_json":"https://pith.science/pith/BE53Y3SPS4PO2EM2OBDAQZ3ZVI.json","view_paper":"https://pith.science/paper/BE53Y3SP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1908.01695&json=true","fetch_graph":"https://pith.science/api/pith-number/BE53Y3SPS4PO2EM2OBDAQZ3ZVI/graph.json","fetch_events":"https://pith.science/api/pith-number/BE53Y3SPS4PO2EM2OBDAQZ3ZVI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BE53Y3SPS4PO2EM2OBDAQZ3ZVI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BE53Y3SPS4PO2EM2OBDAQZ3ZVI/action/storage_attestation","attest_author":"https://pith.science/pith/BE53Y3SPS4PO2EM2OBDAQZ3ZVI/action/author_attestation","sign_citation":"https://pith.science/pith/BE53Y3SPS4PO2EM2OBDAQZ3ZVI/action/citation_signature","submit_replication":"https://pith.science/pith/BE53Y3SPS4PO2EM2OBDAQZ3ZVI/action/replication_record"}},"created_at":"2026-07-05T00:52:27.165696+00:00","updated_at":"2026-07-05T00:52:27.165696+00:00"}