{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:XAQ3YI6WF6HMOBBXE77PED4W65","short_pith_number":"pith:XAQ3YI6W","schema_version":"1.0","canonical_sha256":"b821bc23d62f8ec7043727fef20f96f776b3de19100fdbf8a7b40787b18b4392","source":{"kind":"arxiv","id":"2202.13985","version":2},"attestation_state":"computed","paper":{"title":"The dangers in algorithms learning humans' values and irrationalities","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Rebecca Gorman, Stuart Armstrong","submitted_at":"2022-02-28T17:41:39Z","abstract_excerpt":"For an artificial intelligence (AI) to be aligned with human values (or human preferences), it must first learn those values. AI systems that are trained on human behavior, risk miscategorising human irrationalities as human values -- and then optimising for these irrationalities. Simply learning human values still carries risks: AI learning them will inevitably also gain information on human irrationalities and human behaviour/policy. Both of these can be dangerous: knowing human policy allows an AI to become generically more powerful (whether it is partially aligned or not aligned at all), w"},"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":"2202.13985","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2022-02-28T17:41:39Z","cross_cats_sorted":[],"title_canon_sha256":"77e8e8209ca38bc88978f6522315faaf5b3b1fc1004b0d496d9352a3fe0aaac6","abstract_canon_sha256":"d99b19b75e93e113b56810407fe28355767c909675687e104a6794da983a8b4f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:01:12.911149Z","signature_b64":"7XN5SsSWw9RE2S9xjYOlL1/YM2InZNODUEhZq1hzIRG18HhE7tYl4C5IoWFsUW7N7PjoS8RO+xkqVcYyS48pCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b821bc23d62f8ec7043727fef20f96f776b3de19100fdbf8a7b40787b18b4392","last_reissued_at":"2026-07-05T04:01:12.910696Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:01:12.910696Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The dangers in algorithms learning humans' values and irrationalities","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Rebecca Gorman, Stuart Armstrong","submitted_at":"2022-02-28T17:41:39Z","abstract_excerpt":"For an artificial intelligence (AI) to be aligned with human values (or human preferences), it must first learn those values. AI systems that are trained on human behavior, risk miscategorising human irrationalities as human values -- and then optimising for these irrationalities. Simply learning human values still carries risks: AI learning them will inevitably also gain information on human irrationalities and human behaviour/policy. Both of these can be dangerous: knowing human policy allows an AI to become generically more powerful (whether it is partially aligned or not aligned at all), w"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.13985","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/2202.13985/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":"2202.13985","created_at":"2026-07-05T04:01:12.910756+00:00"},{"alias_kind":"arxiv_version","alias_value":"2202.13985v2","created_at":"2026-07-05T04:01:12.910756+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.13985","created_at":"2026-07-05T04:01:12.910756+00:00"},{"alias_kind":"pith_short_12","alias_value":"XAQ3YI6WF6HM","created_at":"2026-07-05T04:01:12.910756+00:00"},{"alias_kind":"pith_short_16","alias_value":"XAQ3YI6WF6HMOBBX","created_at":"2026-07-05T04:01:12.910756+00:00"},{"alias_kind":"pith_short_8","alias_value":"XAQ3YI6W","created_at":"2026-07-05T04:01:12.910756+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.07269","citing_title":"Datasets for Navigating Sensitive Topics in Recommendation Systems","ref_index":17,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XAQ3YI6WF6HMOBBXE77PED4W65","json":"https://pith.science/pith/XAQ3YI6WF6HMOBBXE77PED4W65.json","graph_json":"https://pith.science/api/pith-number/XAQ3YI6WF6HMOBBXE77PED4W65/graph.json","events_json":"https://pith.science/api/pith-number/XAQ3YI6WF6HMOBBXE77PED4W65/events.json","paper":"https://pith.science/paper/XAQ3YI6W"},"agent_actions":{"view_html":"https://pith.science/pith/XAQ3YI6WF6HMOBBXE77PED4W65","download_json":"https://pith.science/pith/XAQ3YI6WF6HMOBBXE77PED4W65.json","view_paper":"https://pith.science/paper/XAQ3YI6W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2202.13985&json=true","fetch_graph":"https://pith.science/api/pith-number/XAQ3YI6WF6HMOBBXE77PED4W65/graph.json","fetch_events":"https://pith.science/api/pith-number/XAQ3YI6WF6HMOBBXE77PED4W65/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XAQ3YI6WF6HMOBBXE77PED4W65/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XAQ3YI6WF6HMOBBXE77PED4W65/action/storage_attestation","attest_author":"https://pith.science/pith/XAQ3YI6WF6HMOBBXE77PED4W65/action/author_attestation","sign_citation":"https://pith.science/pith/XAQ3YI6WF6HMOBBXE77PED4W65/action/citation_signature","submit_replication":"https://pith.science/pith/XAQ3YI6WF6HMOBBXE77PED4W65/action/replication_record"}},"created_at":"2026-07-05T04:01:12.910756+00:00","updated_at":"2026-07-05T04:01:12.910756+00:00"}