{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004578997,0.0001105617,0.0001807402,0.0001896534,0.00006442559,0.00003353331,0.00005701035,0.0001115923,0.0003729124],"category_scores_gemma":[0.00007472645,0.0001004367,0.00003033734,0.000315094,0.00009564005,0.0001047226,0.00003599747,0.0003148306,0.0003556368],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001434176,"about_ca_system_score_gemma":0.0003184249,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001049636,"about_ca_topic_score_gemma":0.00003487372,"domain_scores_codex":[0.9989218,0.00004002375,0.0003734356,0.0003020956,0.0001013854,0.0002612868],"domain_scores_gemma":[0.9994969,0.0001551791,0.00002370992,0.0001622909,0.00005551953,0.0001063371],"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":[0.00007306873,0.0002678522,0.123682,0.0006932523,0.00001922063,0.0001438933,0.006438903,0.00002085102,0.001023161,0.009176306,0.0003046586,0.8581569],"study_design_scores_gemma":[0.0001016715,0.003934917,0.610836,0.003383986,0.0001611035,0.001332222,0.01570516,0.201001,0.07536428,0.03242606,0.05452659,0.001227012],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9826973,0.001989797,0.0003588849,0.003007348,0.001327167,0.0002765169,0.000001524018,0.00004993303,0.01029153],"genre_scores_gemma":[0.9974421,0.0003669578,0.0005336488,0.00008775001,0.0005079767,0.00005531491,0.00001074529,0.00001510044,0.0009804387],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8569298,"threshold_uncertainty_score":0.457111,"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."}}