{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ZAOW34VWG54ZGA6COLFAMSHBNU","short_pith_number":"pith:ZAOW34VW","schema_version":"1.0","canonical_sha256":"c81d6df2b637799303c272ca0648e16d194c6d0d8f7d4c26ccce8da2fec23d8a","source":{"kind":"arxiv","id":"2303.09387","version":3},"attestation_state":"computed","paper":{"title":"Characterizing Manipulation from AI Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CY","authors_text":"Alan Chan, David Krueger, Henry Ashton, Micah Carroll","submitted_at":"2023-03-16T15:19:21Z","abstract_excerpt":"Manipulation is a common concern in many domains, such as social media, advertising, and chatbots. As AI systems mediate more of our interactions with the world, it is important to understand the degree to which AI systems might manipulate humans without the intent of the system designers. Our work clarifies challenges in defining and measuring manipulation in the context of AI systems. Firstly, we build upon prior literature on manipulation from other fields and characterize the space of possible notions of manipulation, which we find to depend upon the concepts of incentives, intent, harm, a"},"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":"2303.09387","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CY","submitted_at":"2023-03-16T15:19:21Z","cross_cats_sorted":[],"title_canon_sha256":"e83c7dde55a37965621b0c43a73deb8bd7b156a4a930a0a7aeb4e143008499e7","abstract_canon_sha256":"9ab806476642e881861d6d4da42bf728725cce3a9824a6dc9240ff6263368a7a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:06:24.612652Z","signature_b64":"SkskfnDfnYaXHJjTLr9Pm851NmE86+mHCtRF5cRWnOWh6XrpLWj8WOPSkoFF2J9JMafFVzHnXI2lNT4mIRr+BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c81d6df2b637799303c272ca0648e16d194c6d0d8f7d4c26ccce8da2fec23d8a","last_reissued_at":"2026-07-05T07:06:24.612121Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:06:24.612121Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Characterizing Manipulation from AI Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CY","authors_text":"Alan Chan, David Krueger, Henry Ashton, Micah Carroll","submitted_at":"2023-03-16T15:19:21Z","abstract_excerpt":"Manipulation is a common concern in many domains, such as social media, advertising, and chatbots. As AI systems mediate more of our interactions with the world, it is important to understand the degree to which AI systems might manipulate humans without the intent of the system designers. Our work clarifies challenges in defining and measuring manipulation in the context of AI systems. Firstly, we build upon prior literature on manipulation from other fields and characterize the space of possible notions of manipulation, which we find to depend upon the concepts of incentives, intent, harm, a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.09387","kind":"arxiv","version":3},"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/2303.09387/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":"2303.09387","created_at":"2026-07-05T07:06:24.612188+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.09387v3","created_at":"2026-07-05T07:06:24.612188+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.09387","created_at":"2026-07-05T07:06:24.612188+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZAOW34VWG54Z","created_at":"2026-07-05T07:06:24.612188+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZAOW34VWG54ZGA6C","created_at":"2026-07-05T07:06:24.612188+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZAOW34VW","created_at":"2026-07-05T07:06:24.612188+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.08066","citing_title":"Persuasion Attacks Can Decrease Effectiveness of CoT Monitoring","ref_index":83,"is_internal_anchor":true},{"citing_arxiv_id":"2606.05330","citing_title":"A Model of Multi-turn Human Persuadability Using Probabilistic Belief Tracing","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10711","citing_title":"The Agentic Web Requires New Normative Infrastructure","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2411.06837","citing_title":"Persuasion with Large Language Models: A Survey of Empirical Evidence, Study Methodologies, and Ethical Implications","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2401.05561","citing_title":"TrustLLM: Trustworthiness in Large Language Models","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15418","citing_title":"Towards A Framework for Levels of Anthropomorphic Deception in Robots and AI","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZAOW34VWG54ZGA6COLFAMSHBNU","json":"https://pith.science/pith/ZAOW34VWG54ZGA6COLFAMSHBNU.json","graph_json":"https://pith.science/api/pith-number/ZAOW34VWG54ZGA6COLFAMSHBNU/graph.json","events_json":"https://pith.science/api/pith-number/ZAOW34VWG54ZGA6COLFAMSHBNU/events.json","paper":"https://pith.science/paper/ZAOW34VW"},"agent_actions":{"view_html":"https://pith.science/pith/ZAOW34VWG54ZGA6COLFAMSHBNU","download_json":"https://pith.science/pith/ZAOW34VWG54ZGA6COLFAMSHBNU.json","view_paper":"https://pith.science/paper/ZAOW34VW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.09387&json=true","fetch_graph":"https://pith.science/api/pith-number/ZAOW34VWG54ZGA6COLFAMSHBNU/graph.json","fetch_events":"https://pith.science/api/pith-number/ZAOW34VWG54ZGA6COLFAMSHBNU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZAOW34VWG54ZGA6COLFAMSHBNU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZAOW34VWG54ZGA6COLFAMSHBNU/action/storage_attestation","attest_author":"https://pith.science/pith/ZAOW34VWG54ZGA6COLFAMSHBNU/action/author_attestation","sign_citation":"https://pith.science/pith/ZAOW34VWG54ZGA6COLFAMSHBNU/action/citation_signature","submit_replication":"https://pith.science/pith/ZAOW34VWG54ZGA6COLFAMSHBNU/action/replication_record"}},"created_at":"2026-07-05T07:06:24.612188+00:00","updated_at":"2026-07-05T07:06:24.612188+00:00"}