{"id":"W4394785742","doi":"10.1038/s41433-024-03067-4","title":"Google Gemini and Bard artificial intelligence chatbot performance in ophthalmology knowledge assessment","year":2024,"lang":"en","type":"article","venue":"Eye","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":51,"is_retracted":false,"has_abstract":false,"ca_institutions":"St. Michael's Hospital; Health Sciences Centre; Sunnybrook Health Science Centre; Hospital for Sick Children; McMaster University; University of Toronto","funders":"","keywords":"Chatbot; Neuro-ophthalmology; Optometry; Ophthalmology; Medicine; Computer science; Artificial intelligence; Glaucoma","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":[0.003912915,0.0008797839,0.0008234653,0.001771986,0.0006942476,0.00269029,0.001032605,0.002083567,0.01020804],"category_scores_gemma":[0.03911095,0.0003155906,0.0005018148,0.000841942,0.0004417963,0.003956508,0.002657895,0.001488634,0.006330496],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009461485,"about_ca_system_score_gemma":0.001309298,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01701687,"about_ca_topic_score_gemma":0.01925883,"domain_scores_codex":[0.9962097,0.001557354,0.0003567543,0.0005462843,0.000967693,0.0003621986],"domain_scores_gemma":[0.9650857,0.02720624,0.0009647041,0.00156986,0.002933607,0.002239902],"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":[0.03068614,0.00626976,0.1375023,0.004117884,0.0007674143,0.0008145091,0.007157004,0.01394614,0.0139505,0.003038986,0.1468851,0.6348642],"study_design_scores_gemma":[0.0019142,0.01443986,0.4779096,0.001416244,0.001444713,0.001321008,0.01207795,0.3354483,0.04108414,0.00935394,0.1026914,0.0008986653],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9504442,0.001237314,0.002852891,0.0008908518,0.0005726736,0.0002251034,0.004108143,0.01277343,0.02689549],"genre_scores_gemma":[0.9717249,0.0003010339,0.00646252,0.000417986,0.00009640579,0.0001358065,0.006496936,0.0004837504,0.01388059],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01701687,"threshold_uncertainty_score":0.03414929,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1933820781356063,"score_gpt":0.4910439300606607,"score_spread":0.2976618519250543,"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."}}