{"paper":{"title":"Sources of Truth: A Multi-Platform, Multilingual Audit of Citations in AI Mental Health Information Queries","arxiv_id":"2609.00319","primary_cat":"cs.CY","submitted_at":"2026-08-31T20:09:28+00:00","version_compiled":"v1"},"notice":"Every quoted sentence is the authors' text from the arXiv listing abstract. Labels and probabilities are the outputs of classifier-1.13.0 on the question bank abstract-p1-20260923, exactly as returned. No sentence in this record was written by a language model.","schema":"pith.compiled-record/v0","status":"compiled","abstract":{"source":"arXiv listing abstract v1","sentences":[{"i":0,"p":1.0,"id":"s0","own":false,"role":"background","text":"Online health information seeking is shifting from keyword search, where users consider a ranked list of links, to conversational systems that compose a single answer and curate its citations.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.05,"main_claim_p":0.03,"quantitative":false,"quantitative_p":0.33,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.0,"background":1.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":1,"p":0.92,"id":"s1","own":false,"role":"gap","text":"Source evaluation therefore passes from user to platform, yet what these systems surface is poorly characterized.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.05,"main_claim_p":0.06,"quantitative":false,"quantitative_p":0.06,"role_probabilities":{"gap":0.92,"method":0.0,"result":0.0,"background":0.08,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":2,"p":0.95,"id":"s2","own":true,"role":"method","text":"We audited three free consumer products (ChatGPT, Perplexity, Google AI Overview) on twenty English mental health questions under two prompt conditions, with a subset of three also translated into six further languages of varying resource tiers.","hedge":{"p":0.94,"label":"asserted"},"own_p":0.97,"main_claim_p":0.03,"quantitative":true,"quantitative_p":0.95,"role_probabilities":{"gap":0.0,"method":0.95,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.05,"interpretation":0.0}},{"i":3,"p":0.98,"id":"s3","own":true,"role":"method","text":"We recorded 15,942 citations across 1,140 responses and 1,713 unique domains, then classified every citation with a nine-category organizational typology applied by a deterministic classifier validated against human coding.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.98,"main_claim_p":0.05,"quantitative":true,"quantitative_p":0.98,"role_probabilities":{"gap":0.0,"method":0.98,"result":0.01,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.01,"interpretation":0.0}},{"i":4,"p":1.0,"id":"s4","own":true,"role":"result","text":"Citations were heavily concentrated: the ten most-cited domains accounted for 43.6% of English citations, and government, commercial health, and academic sources were closely matched at roughly 22% each.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.97,"main_claim_p":0.15,"quantitative":true,"quantitative_p":0.99,"role_probabilities":{"gap":0.0,"method":0.0,"result":1.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":5,"p":1.0,"id":"s5","own":true,"role":"result","text":"Platforms differed little in typical citation volume but sharply in consistency and in the source types they favored.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.95,"main_claim_p":0.22,"quantitative":true,"quantitative_p":0.9,"role_probabilities":{"gap":0.0,"method":0.0,"result":1.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":6,"p":1.0,"id":"s6","own":true,"role":"result","text":"Explicitly requesting sources shifted composition only modestly.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.94,"main_claim_p":0.05,"quantitative":true,"quantitative_p":0.6,"role_probabilities":{"gap":0.0,"method":0.0,"result":1.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":7,"p":1.0,"id":"s7","own":true,"role":"result","text":"Non-English queries surfaced fewer citations and were routed to language-appropriate resources at significantly lower rates.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.95,"main_claim_p":0.13,"quantitative":true,"quantitative_p":0.88,"role_probabilities":{"gap":0.0,"method":0.0,"result":1.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":8,"p":0.95,"id":"s8","own":true,"role":"contribution","text":"We release the typology, classifier, and annotated corpus as reusable instruments for auditing generative health search.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.93,"main_claim_p":0.06,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.01,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.04,"contribution":0.95,"interpretation":0.0}}]},"compiled":{"main_claim":{"p":0.85,"text":"Platforms differed little in typical citation volume but sharply in consistency and in the source types they favored.","location":"abstract s5","hedge_strength":{"p":1.0,"label":"asserted"}},"provenance":{"route":{"via":"TypeSafe API","model":"classifier-1.13.0","endpoint":"https://api.typesafe.ai/v1/systemone","provider":"TypeSafe"},"packing":"one request per paper","requests":1,"input_tokens":6905,"cost_usd_reported":0.0},"compiled_at":"2026-09-23","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.05,"states_limitation":0.49,"human_participants":0.17,"controlled_comparison":0.13,"randomized_assignment":0.03,"states_quantitative_result":0.99,"compares_with_prior_results":0.25,"mentions_code_or_data_availability":0.95},"evidential_basis":{"p":0.98,"value":"observation or experiment","probabilities":{"argument":0.0,"literature":0.0,"proof_or_derivation":0.0,"analysis_of_existing_data":0.02,"observation_or_experiment":0.98,"simulation_or_computation":0.0}},"contribution_type":{"p":0.88,"value":"empirical study","probabilities":{"review":0.0,"method_or_tool":0.01,"empirical_study":0.88,"simulation_study":0.0,"theoretical_result":0.0,"dataset_or_benchmark":0.11,"position_or_commentary":0.0}},"contribution_statement":{"p":0.87,"text":"We audited three free consumer products (ChatGPT, Perplexity, Google AI Overview) on twenty English mental health questions under two prompt conditions, with a subset of three also translated into six further languages of varying resource tiers.","location":"abstract s2"}},"coverage":{"tier":"abstract","levels_run":["P abstract"],"not_examined":["full text","figures and tables","references and citing sentences","relations to other papers"],"sources_read":["arXiv listing abstract v1"],"levels_not_run":["S sections","T body sentences","C citations","R cross-paper relations"],"sentences_examined":9,"sentences_in_abstract":9},"measured_accuracy":{"url":"https://pith.science/accuracy","measured_on":"2026-10-02","method":"two independent model families labelled a fresh random sample blind; the figure is how often the published label matches their consensus","sentence_role_agreement":0.92,"main_claim_agreement":0.908,"typed_citation_agreement":0.933,"typed_citation_agreement_95_interval":[0.917,0.945],"review_comments_accurate":0.758,"review_comments_accurate_95_interval":[0.684,0.82],"review_caught_admitted_error_found_or_related":0.448}}