{"id":"W4409283601","doi":"10.1038/s41586-025-08866-7","title":"Towards conversational diagnostic artificial intelligence","year":2025,"lang":"en","type":"article","venue":"Nature","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":246,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"JSS Academy of Higher Education and Research; DeepMind","keywords":"Milestone; Empathy; Objective structured clinical examination; Excellence; Diagnostic accuracy; Scale (ratio); Medical education; Primary care; Medical history; Artificial intelligence; Patient care; Psychology; Computer science; Medicine; Family medicine; Nursing; Radiology; Social psychology","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001747425,0.00008532376,0.0001253929,0.0001346325,0.0000952103,0.00002008241,0.00007735087,0.0007037821,0.0004978464],"category_scores_gemma":[0.002470794,0.00007688484,0.00005723308,0.0004039258,0.00006633803,0.00005362087,0.00001829852,0.001220617,0.0001938686],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001181942,"about_ca_system_score_gemma":0.0005596678,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001017308,"about_ca_topic_score_gemma":0.00003591656,"domain_scores_codex":[0.9991593,0.00002153354,0.0002541079,0.0001997594,0.0001986193,0.0001667141],"domain_scores_gemma":[0.9990113,0.0004021653,0.00003469039,0.0001788847,0.00029745,0.0000754967],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001577605,0.0002263143,0.01346995,0.0001548206,0.00004479413,0.00001833179,0.0007616419,0.00002718582,0.0005155319,0.2457812,0.02169492,0.7171476],"study_design_scores_gemma":[0.00006920644,0.0003689093,0.0997489,0.0007360104,0.0002412203,0.00004811395,0.004986898,0.003781847,0.2650241,0.5131837,0.1113505,0.0004605952],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6634771,0.0052707,0.02646857,0.2461778,0.009393648,0.001522228,0.00002484525,0.0003263194,0.0473388],"genre_scores_gemma":[0.9898384,0.00009143584,0.0007774807,0.008013143,0.0004771257,0.00002417223,0.00004572974,0.000005833882,0.0007266956],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.716687,"threshold_uncertainty_score":0.5451069,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07912451454278564,"score_gpt":0.4313034769996503,"score_spread":0.3521789624568646,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}