{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:4YX2EX5KUNNJF7MTUZQPVQMT4T","short_pith_number":"pith:4YX2EX5K","schema_version":"1.0","canonical_sha256":"e62fa25faaa35a92fd93a660fac193e4f1b04308ad46cf597fc494bd46c05c49","source":{"kind":"arxiv","id":"2310.16048","version":1},"attestation_state":"computed","paper":{"title":"AI Alignment and Social Choice: Fundamental Limitations and Policy Implications","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.CY","cs.HC","cs.LG"],"primary_cat":"cs.AI","authors_text":"Abhilash Mishra","submitted_at":"2023-10-24T17:59:04Z","abstract_excerpt":"Aligning AI agents to human intentions and values is a key bottleneck in building safe and deployable AI applications. But whose values should AI agents be aligned with? Reinforcement learning with human feedback (RLHF) has emerged as the key framework for AI alignment. RLHF uses feedback from human reinforcers to fine-tune outputs; all widely deployed large language models (LLMs) use RLHF to align their outputs to human values. It is critical to understand the limitations of RLHF and consider policy challenges arising from these limitations. In this paper, we investigate a specific challenge "},"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":"2310.16048","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2023-10-24T17:59:04Z","cross_cats_sorted":["cs.CL","cs.CY","cs.HC","cs.LG"],"title_canon_sha256":"7704c696ecec7cea33520db3cb682476e8a9b3e334c7120e10981bd5cd7ecf54","abstract_canon_sha256":"f4b8c7bb6063350e6ca28caaadce9e670fd71f52857520d792a7294f596a59f3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:04:34.360938Z","signature_b64":"2GU8l3WP72WVRAtrOArJdVD93OuKrrNulHGKJh8zCMGydWBMPaEQKpWFpBlb/2piQfitBPzKLVJHROy2i1WPCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e62fa25faaa35a92fd93a660fac193e4f1b04308ad46cf597fc494bd46c05c49","last_reissued_at":"2026-07-05T07:04:34.360493Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:04:34.360493Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AI Alignment and Social Choice: Fundamental Limitations and Policy Implications","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.CY","cs.HC","cs.LG"],"primary_cat":"cs.AI","authors_text":"Abhilash Mishra","submitted_at":"2023-10-24T17:59:04Z","abstract_excerpt":"Aligning AI agents to human intentions and values is a key bottleneck in building safe and deployable AI applications. But whose values should AI agents be aligned with? Reinforcement learning with human feedback (RLHF) has emerged as the key framework for AI alignment. RLHF uses feedback from human reinforcers to fine-tune outputs; all widely deployed large language models (LLMs) use RLHF to align their outputs to human values. It is critical to understand the limitations of RLHF and consider policy challenges arising from these limitations. In this paper, we investigate a specific challenge "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.16048","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/2310.16048/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":"2310.16048","created_at":"2026-07-05T07:04:34.360549+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.16048v1","created_at":"2026-07-05T07:04:34.360549+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.16048","created_at":"2026-07-05T07:04:34.360549+00:00"},{"alias_kind":"pith_short_12","alias_value":"4YX2EX5KUNNJ","created_at":"2026-07-05T07:04:34.360549+00:00"},{"alias_kind":"pith_short_16","alias_value":"4YX2EX5KUNNJF7MT","created_at":"2026-07-05T07:04:34.360549+00:00"},{"alias_kind":"pith_short_8","alias_value":"4YX2EX5K","created_at":"2026-07-05T07:04:34.360549+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21550","citing_title":"AI Alignment From Social Choice Perspectives","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2504.12501","citing_title":"Reinforcement Learning from Human Feedback","ref_index":218,"is_internal_anchor":false},{"citing_arxiv_id":"2602.13372","citing_title":"MoralityGym: A Benchmark for Evaluating Hierarchical Moral Alignment in Sequential Decision-Making Agents","ref_index":66,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11865","citing_title":"Variance-aware Reward Modeling with Anchor Guidance","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07724","citing_title":"Curated Synthetic Data Doesn't Have to Collapse: A Theoretical Study of Generative Retraining with Pluralistic Preferences","ref_index":80,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02379","citing_title":"Fair Agents: Balancing Multistakeholder Alignment in Multi-Agent Personalization Systems","ref_index":21,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4YX2EX5KUNNJF7MTUZQPVQMT4T","json":"https://pith.science/pith/4YX2EX5KUNNJF7MTUZQPVQMT4T.json","graph_json":"https://pith.science/api/pith-number/4YX2EX5KUNNJF7MTUZQPVQMT4T/graph.json","events_json":"https://pith.science/api/pith-number/4YX2EX5KUNNJF7MTUZQPVQMT4T/events.json","paper":"https://pith.science/paper/4YX2EX5K"},"agent_actions":{"view_html":"https://pith.science/pith/4YX2EX5KUNNJF7MTUZQPVQMT4T","download_json":"https://pith.science/pith/4YX2EX5KUNNJF7MTUZQPVQMT4T.json","view_paper":"https://pith.science/paper/4YX2EX5K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.16048&json=true","fetch_graph":"https://pith.science/api/pith-number/4YX2EX5KUNNJF7MTUZQPVQMT4T/graph.json","fetch_events":"https://pith.science/api/pith-number/4YX2EX5KUNNJF7MTUZQPVQMT4T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4YX2EX5KUNNJF7MTUZQPVQMT4T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4YX2EX5KUNNJF7MTUZQPVQMT4T/action/storage_attestation","attest_author":"https://pith.science/pith/4YX2EX5KUNNJF7MTUZQPVQMT4T/action/author_attestation","sign_citation":"https://pith.science/pith/4YX2EX5KUNNJF7MTUZQPVQMT4T/action/citation_signature","submit_replication":"https://pith.science/pith/4YX2EX5KUNNJF7MTUZQPVQMT4T/action/replication_record"}},"created_at":"2026-07-05T07:04:34.360549+00:00","updated_at":"2026-07-05T07:04:34.360549+00:00"}