{"id":"W4404717881","doi":"10.1186/s12909-024-06309-x","title":"Performance of ChatGPT and Bard on the medical licensing examinations varies across different cultures: a comparison study","year":2024,"lang":"en","type":"article","venue":"BMC Medical Education","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":42,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"United States Medical Licensing Examination; Medicine; Asian americans; Family medicine; Significant difference; Medical education; Medical physics; Medical school; Internal medicine; Political science; Ethnic group","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001013576,0.0001154967,0.0002153752,0.00007203437,0.000216858,0.00004559609,0.0001002233,0.0001399064,0.0003548637],"category_scores_gemma":[0.004337177,0.00006758189,0.0000373364,0.000245465,0.0002326606,0.00006411207,0.00003419773,0.0004011851,0.0000253014],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007985804,"about_ca_system_score_gemma":0.001783994,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003684845,"about_ca_topic_score_gemma":0.0006545019,"domain_scores_codex":[0.9980522,0.0001531525,0.0004778075,0.0002383523,0.0008984134,0.0001800208],"domain_scores_gemma":[0.9980509,0.001221031,0.00006468738,0.0002214414,0.0001685654,0.0002734099],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0001587962,0.003845347,0.3366652,0.001445625,0.00007661113,0.000003682557,0.1160254,0.00000322482,0.00006035755,0.005906052,0.00498161,0.5308281],"study_design_scores_gemma":[0.0002146285,0.002065527,0.718247,0.00565758,0.0002103903,0.0001149959,0.1806308,0.08371172,0.002669462,0.0005250768,0.005666936,0.0002859217],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9834162,0.0007254523,0.000252935,0.01346307,0.001316593,0.0005428707,0.000001187579,0.00004117965,0.0002405098],"genre_scores_gemma":[0.997887,0.0002937768,0.00004713472,0.0007637056,0.0006814069,0.00009771894,0.00002178073,0.00001095702,0.0001965624],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5305421,"threshold_uncertainty_score":0.519232,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1546363183540944,"score_gpt":0.4981344811457172,"score_spread":0.3434981627916227,"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."}}