{"id":"W4364378939","doi":"10.2196/46599","title":"Trialling a Large Language Model (ChatGPT) in General Practice With the Applied Knowledge Test: Observational Study Demonstrating Opportunities and Limitations in Primary Care","year":2023,"lang":"en","type":"article","venue":"JMIR Medical Education","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":192,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Generative grammar; Observational study; Transformer; Computer science; Test (biology); Knowledge management; Artificial intelligence; Medicine; Engineering; Pathology","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001090421,0.0001018426,0.0001612971,0.000205701,0.0001406792,0.0000290594,0.0000675406,0.00009474142,0.00001042847],"category_scores_gemma":[0.002041741,0.00007639001,0.00001490156,0.0005928779,0.00006024364,0.00015822,0.00002435576,0.0003521895,0.000006508769],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001895873,"about_ca_system_score_gemma":0.004621107,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002437383,"about_ca_topic_score_gemma":0.003047212,"domain_scores_codex":[0.998693,0.0001132468,0.0003954498,0.0002194689,0.0003694791,0.0002093026],"domain_scores_gemma":[0.9980509,0.001366824,0.00009884246,0.0001468402,0.0002037678,0.0001328328],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"qualitative","study_design_scores_codex":[0.0003127772,0.003443666,0.1014426,0.0004694346,0.00002189009,0.00002478499,0.5672907,0.0002701576,0.0001790345,0.003636131,0.001099389,0.3218095],"study_design_scores_gemma":[0.0005225987,0.0002148685,0.1445086,0.0003033596,0.00003655908,0.00001459553,0.81941,0.03387397,0.00002048364,0.0001866629,0.0007756216,0.0001326838],"study_design_candidate":"qualitative","study_design_consensus":"qualitative","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9747011,0.0003307335,0.00007565234,0.01857632,0.0001130636,0.00141945,0.000002448312,0.00004253188,0.004738695],"genre_scores_gemma":[0.9953114,0.0001439378,0.0006529799,0.001724363,0.0003595508,0.001139006,0.0002636854,0.00001506674,0.0003900147],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3216768,"threshold_uncertainty_score":0.8197645,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3570125126813295,"score_gpt":0.4801588478092852,"score_spread":0.1231463351279557,"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."}}