{"id":"W4392098566","doi":"10.1016/j.jcjo.2024.01.011","title":"Performance of three artificial intelligence chatbots on Ophthalmic Knowledge Assessment Program materials","year":2024,"lang":"en","type":"letter","venue":"Canadian Journal of Ophthalmology","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":8,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"","keywords":"Test (biology); Computer science; Artificial intelligence; Set (abstract data type); Graphics; Psychology","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.009282552,0.001394404,0.001164284,0.001766091,0.003181045,0.003291165,0.002308018,0.003823029,0.01408187],"category_scores_gemma":[0.06351389,0.0009751386,0.0006407116,0.0007472499,0.0006194627,0.002242265,0.003243827,0.00211433,0.007142562],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002158842,"about_ca_system_score_gemma":0.002588473,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01040409,"about_ca_topic_score_gemma":0.01378686,"domain_scores_codex":[0.9922748,0.00384108,0.0006155474,0.0008725977,0.001510726,0.0008851119],"domain_scores_gemma":[0.8938122,0.08381591,0.002276093,0.004145359,0.009109552,0.006840891],"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":[0.03473037,0.07862045,0.09090643,0.001834563,0.0004939661,0.005659303,0.04573281,0.00839589,0.06103257,0.001777024,0.05985402,0.6109626],"study_design_scores_gemma":[0.005851899,0.07744928,0.3898674,0.001157855,0.001245201,0.00238006,0.04359325,0.2386948,0.1663533,0.004659656,0.06733768,0.001409613],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9840703,0.00005296882,0.002795676,0.000497981,0.0002332485,0.001143731,0.0003457011,0.002763069,0.008097365],"genre_scores_gemma":[0.9583821,0.00006023814,0.01505054,0.0006832519,0.0001006611,0.0009575203,0.001348819,0.0003673293,0.02304956],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01408187,"threshold_uncertainty_score":0.0490914,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2206070772845467,"score_gpt":0.4441734811624296,"score_spread":0.2235664038778829,"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."}}