{"id":"W4388694463","doi":"10.1007/s00264-023-06034-y","title":"Does Google’s Bard Chatbot perform better than ChatGPT on the European hand surgery exam?","year":2023,"lang":"en","type":"article","venue":"International Orthopaedics","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":22,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Medicine; Chatbot; Homogeneous; Multiple choice; Kappa; Surgery; Medical education; Natural language processing; Linguistics; Computer science; Significant difference; Internal medicine; Mathematics","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001452994,0.0001607382,0.0001703635,0.0002625187,0.0002497079,0.00008975717,0.0001962835,0.00007066581,0.0008599923],"category_scores_gemma":[0.0008258711,0.00008708996,0.0001864042,0.0002854198,0.000137138,0.000113585,0.00006253782,0.0003375517,0.002143515],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009899891,"about_ca_system_score_gemma":0.0001380232,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001458301,"about_ca_topic_score_gemma":0.00008500787,"domain_scores_codex":[0.9981961,0.0001122788,0.0004552008,0.0002900216,0.0006453321,0.0003011301],"domain_scores_gemma":[0.9983523,0.0007948828,0.0001254971,0.0003246266,0.0002622864,0.0001403575],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001367309,0.0001321744,0.5851498,0.00004639767,0.0001182561,0.0001450895,0.002524835,0.00001512009,0.0002824756,0.001286262,0.04667119,0.3634917],"study_design_scores_gemma":[0.0001105312,0.0001645669,0.6526359,0.0004932237,0.00004900775,0.00005089993,0.001883301,0.002011074,0.006930609,0.001850529,0.3335021,0.000318288],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8982588,0.00002407146,0.00005602928,0.08576559,0.004888941,0.0002277353,0.00001831523,0.0001218894,0.01063859],"genre_scores_gemma":[0.9844217,0.0003233369,0.00002239949,0.005759369,0.004350245,0.00003876188,0.0001910705,0.00003966251,0.004853438],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3631734,"threshold_uncertainty_score":0.9986334,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1287397648107904,"score_gpt":0.3624818366539341,"score_spread":0.2337420718431437,"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."}}