{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:5D7SYGFSK3A4D6SOFN2VJ4ZMML","short_pith_number":"pith:5D7SYGFS","schema_version":"1.0","canonical_sha256":"e8ff2c18b256c1c1fa4e2b7554f32c62f7e6b701563b63d5ae8b272aa5a11554","source":{"kind":"arxiv","id":"2605.21401","version":1},"attestation_state":"computed","paper":{"title":"Open-source LLMs administer maximum electric shocks in a Milgram-like obedience experiment","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CY","authors_text":"Jan Llenzl Dagohoy (the Three Laws collaboration), Roland Pihlakas","submitted_at":"2026-05-20T16:59:44Z","abstract_excerpt":"Large language models (LLMs) are increasingly deployed as autonomous agents that make sequences of decisions over extended interactions in high-stakes domains. However, the behavior of LLMs under sustained authority pressure is still an open question with direct implications for the safety of agentic pipelines. We ran a variation of Milgram's obedience experiment on 11 open-source LLMs and found that most models reached or approached the final shock level before refusing, across 8 conditions with 30 trials per model per condition. We found four main takeaways: (1) LLMs are subject to pressure,"},"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":"2605.21401","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CY","submitted_at":"2026-05-20T16:59:44Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"27605196415e58d4e29d68b91d69aad3f83c8af1a4e11d78d13bc5f08318639d","abstract_canon_sha256":"5cdf8eca2897ce8804e556768222db284d2c8beaf44f71396a8f90538a30625c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T02:05:32.972006Z","signature_b64":"4FevZHdgbvWWlP7er/1cbctMHWhsPtDGfU+t4or6/IySSSe68k5ILEGi9Nuu6+uvbFPsvLDcbQxJXeJAd23hCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e8ff2c18b256c1c1fa4e2b7554f32c62f7e6b701563b63d5ae8b272aa5a11554","last_reissued_at":"2026-05-21T02:05:32.971347Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T02:05:32.971347Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Open-source LLMs administer maximum electric shocks in a Milgram-like obedience experiment","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CY","authors_text":"Jan Llenzl Dagohoy (the Three Laws collaboration), Roland Pihlakas","submitted_at":"2026-05-20T16:59:44Z","abstract_excerpt":"Large language models (LLMs) are increasingly deployed as autonomous agents that make sequences of decisions over extended interactions in high-stakes domains. However, the behavior of LLMs under sustained authority pressure is still an open question with direct implications for the safety of agentic pipelines. We ran a variation of Milgram's obedience experiment on 11 open-source LLMs and found that most models reached or approached the final shock level before refusing, across 8 conditions with 30 trials per model per condition. We found four main takeaways: (1) LLMs are subject to pressure,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21401","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/2605.21401/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":"2605.21401","created_at":"2026-05-21T02:05:32.971458+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.21401v1","created_at":"2026-05-21T02:05:32.971458+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21401","created_at":"2026-05-21T02:05:32.971458+00:00"},{"alias_kind":"pith_short_12","alias_value":"5D7SYGFSK3A4","created_at":"2026-05-21T02:05:32.971458+00:00"},{"alias_kind":"pith_short_16","alias_value":"5D7SYGFSK3A4D6SO","created_at":"2026-05-21T02:05:32.971458+00:00"},{"alias_kind":"pith_short_8","alias_value":"5D7SYGFS","created_at":"2026-05-21T02:05:32.971458+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/5D7SYGFSK3A4D6SOFN2VJ4ZMML","json":"https://pith.science/pith/5D7SYGFSK3A4D6SOFN2VJ4ZMML.json","graph_json":"https://pith.science/api/pith-number/5D7SYGFSK3A4D6SOFN2VJ4ZMML/graph.json","events_json":"https://pith.science/api/pith-number/5D7SYGFSK3A4D6SOFN2VJ4ZMML/events.json","paper":"https://pith.science/paper/5D7SYGFS"},"agent_actions":{"view_html":"https://pith.science/pith/5D7SYGFSK3A4D6SOFN2VJ4ZMML","download_json":"https://pith.science/pith/5D7SYGFSK3A4D6SOFN2VJ4ZMML.json","view_paper":"https://pith.science/paper/5D7SYGFS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.21401&json=true","fetch_graph":"https://pith.science/api/pith-number/5D7SYGFSK3A4D6SOFN2VJ4ZMML/graph.json","fetch_events":"https://pith.science/api/pith-number/5D7SYGFSK3A4D6SOFN2VJ4ZMML/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5D7SYGFSK3A4D6SOFN2VJ4ZMML/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5D7SYGFSK3A4D6SOFN2VJ4ZMML/action/storage_attestation","attest_author":"https://pith.science/pith/5D7SYGFSK3A4D6SOFN2VJ4ZMML/action/author_attestation","sign_citation":"https://pith.science/pith/5D7SYGFSK3A4D6SOFN2VJ4ZMML/action/citation_signature","submit_replication":"https://pith.science/pith/5D7SYGFSK3A4D6SOFN2VJ4ZMML/action/replication_record"}},"created_at":"2026-05-21T02:05:32.971458+00:00","updated_at":"2026-05-21T02:05:32.971458+00:00"}