{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:2CS5RBY35S2JCVKKE7GG4VS2VE","short_pith_number":"pith:2CS5RBY3","schema_version":"1.0","canonical_sha256":"d0a5d8871becb491554a27cc6e565aa93be5ed1e53aabee24f26e033935fe98a","source":{"kind":"arxiv","id":"2507.15907","version":1},"attestation_state":"computed","paper":{"title":"Dual Turing Test: A Framework for Detecting and Mitigating Undetectable AI","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Alberto Messina","submitted_at":"2025-07-21T13:44:28Z","abstract_excerpt":"In this short note, we propose a unified framework that bridges three areas: (1) a flipped perspective on the Turing Test, the \"dual Turing test\", in which a human judge's goal is to identify an AI rather than reward a machine for deception; (2) a formal adversarial classification game with explicit quality constraints and worst-case guarantees; and (3) a reinforcement learning (RL) alignment pipeline that uses an undetectability detector and a set of quality related components in its reward model. We review historical precedents, from inverted and meta-Turing variants to modern supervised rev"},"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":"2507.15907","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-21T13:44:28Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2770ae0356233704604b8a272139611f695465b2b2e544443e4183e1ba0961af","abstract_canon_sha256":"ae27d9b06f4ad3174486dc5dda36d291e7064c09146c6bedf7bc1559c06e2f56"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:40:59.064818Z","signature_b64":"fEFGWbJxDWx23EOL1kygHvXJUe73X9nyzcTzDycWDdBkMoElDyR0tB3WTz8v8TnCxceR/xlSfZNYowwrO2e5Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d0a5d8871becb491554a27cc6e565aa93be5ed1e53aabee24f26e033935fe98a","last_reissued_at":"2026-07-05T11:40:59.064338Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:40:59.064338Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dual Turing Test: A Framework for Detecting and Mitigating Undetectable AI","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Alberto Messina","submitted_at":"2025-07-21T13:44:28Z","abstract_excerpt":"In this short note, we propose a unified framework that bridges three areas: (1) a flipped perspective on the Turing Test, the \"dual Turing test\", in which a human judge's goal is to identify an AI rather than reward a machine for deception; (2) a formal adversarial classification game with explicit quality constraints and worst-case guarantees; and (3) a reinforcement learning (RL) alignment pipeline that uses an undetectability detector and a set of quality related components in its reward model. We review historical precedents, from inverted and meta-Turing variants to modern supervised rev"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.15907","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/2507.15907/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":"2507.15907","created_at":"2026-07-05T11:40:59.064399+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.15907v1","created_at":"2026-07-05T11:40:59.064399+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.15907","created_at":"2026-07-05T11:40:59.064399+00:00"},{"alias_kind":"pith_short_12","alias_value":"2CS5RBY35S2J","created_at":"2026-07-05T11:40:59.064399+00:00"},{"alias_kind":"pith_short_16","alias_value":"2CS5RBY35S2JCVKK","created_at":"2026-07-05T11:40:59.064399+00:00"},{"alias_kind":"pith_short_8","alias_value":"2CS5RBY3","created_at":"2026-07-05T11:40:59.064399+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21844","citing_title":"Inverse Turing Bench: Evaluating Language Models as Judges of Human vs. AI Dialogue","ref_index":58,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2CS5RBY35S2JCVKKE7GG4VS2VE","json":"https://pith.science/pith/2CS5RBY35S2JCVKKE7GG4VS2VE.json","graph_json":"https://pith.science/api/pith-number/2CS5RBY35S2JCVKKE7GG4VS2VE/graph.json","events_json":"https://pith.science/api/pith-number/2CS5RBY35S2JCVKKE7GG4VS2VE/events.json","paper":"https://pith.science/paper/2CS5RBY3"},"agent_actions":{"view_html":"https://pith.science/pith/2CS5RBY35S2JCVKKE7GG4VS2VE","download_json":"https://pith.science/pith/2CS5RBY35S2JCVKKE7GG4VS2VE.json","view_paper":"https://pith.science/paper/2CS5RBY3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.15907&json=true","fetch_graph":"https://pith.science/api/pith-number/2CS5RBY35S2JCVKKE7GG4VS2VE/graph.json","fetch_events":"https://pith.science/api/pith-number/2CS5RBY35S2JCVKKE7GG4VS2VE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2CS5RBY35S2JCVKKE7GG4VS2VE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2CS5RBY35S2JCVKKE7GG4VS2VE/action/storage_attestation","attest_author":"https://pith.science/pith/2CS5RBY35S2JCVKKE7GG4VS2VE/action/author_attestation","sign_citation":"https://pith.science/pith/2CS5RBY35S2JCVKKE7GG4VS2VE/action/citation_signature","submit_replication":"https://pith.science/pith/2CS5RBY35S2JCVKKE7GG4VS2VE/action/replication_record"}},"created_at":"2026-07-05T11:40:59.064399+00:00","updated_at":"2026-07-05T11:40:59.064399+00:00"}