{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003787879,0.0005385554,0.0007700888,0.0008480565,0.0005942675,0.001567399,0.0006859694,0.001839878,0.01363505],"category_scores_gemma":[0.02831996,0.0002840347,0.0008823242,0.0005077411,0.0003992706,0.002399164,0.0008840708,0.0009852251,0.009261601],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005119347,"about_ca_system_score_gemma":0.001050755,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01074583,"about_ca_topic_score_gemma":0.02527932,"domain_scores_codex":[0.998261,0.0007007581,0.0001306979,0.0002474941,0.0003899213,0.0002700886],"domain_scores_gemma":[0.9800119,0.01199866,0.001524568,0.0009177671,0.002347721,0.003199317],"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":[0.02341778,0.004730717,0.3763696,0.001176181,0.001124479,0.0002710263,0.001313605,0.002187018,0.001835762,0.0003588382,0.1511839,0.4360312],"study_design_scores_gemma":[0.002322667,0.01148464,0.9152426,0.0008903604,0.001181503,0.0005921841,0.004067459,0.01504534,0.002713327,0.001537182,0.04463079,0.0002920038],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9612609,0.002651949,0.0008480083,0.007304449,0.002177852,0.0001778304,0.004284431,0.001331326,0.01996325],"genre_scores_gemma":[0.9826696,0.0007095836,0.001680508,0.002353563,0.0004471184,0.00008718414,0.0045304,0.00017815,0.007343851],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01363505,"threshold_uncertainty_score":0.04561377,"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."}}