{"id":"W4392800721","doi":"10.1001/jama.2024.3620","title":"The Limits of Clinician Vigilance as an AI Safety Bulwark","year":2024,"lang":"en","type":"article","venue":"JAMA","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":54,"is_retracted":false,"has_abstract":true,"ca_institutions":"Health Sciences Centre; University of Toronto; Sunnybrook Health Science Centre","funders":"","keywords":"Vigilance (psychology); Medicine; Occupational safety and health; Poison control; Medical emergency; Cognitive psychology; Pathology; Psychology","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":[0.0005143936,0.000057212,0.0001089289,0.00003287457,0.00009931517,0.0000350299,0.00007969813,0.00008529368,0.00007442567],"category_scores_gemma":[0.0004952964,0.00003760864,0.00005178055,0.0001803643,0.0000682313,0.00009629186,0.000008144703,0.0002345002,0.0002845054],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003916342,"about_ca_system_score_gemma":0.0003090143,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004273332,"about_ca_topic_score_gemma":0.0001754195,"domain_scores_codex":[0.9991781,0.00004238828,0.0003360298,0.0001464638,0.0001478356,0.0001491463],"domain_scores_gemma":[0.99905,0.0003951723,0.00003849739,0.0002689144,0.0001597749,0.00008761793],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0002540938,0.00006293837,0.005434304,0.0001414104,0.00002003299,0.00001059126,0.002328581,0.00000336288,0.000373581,0.01317905,0.007864201,0.9703279],"study_design_scores_gemma":[0.00008068526,0.001223601,0.03126327,0.0009140418,0.00006973885,0.0000525187,0.004338826,0.003638051,0.0361913,0.02341623,0.898656,0.0001557195],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8835416,0.003007997,0.0001725073,0.09781462,0.002348777,0.0003102914,0.000005037545,0.00008946705,0.0127097],"genre_scores_gemma":[0.99297,0.0007696147,0.00007816964,0.002965781,0.001227414,0.000009546793,0.000009492193,0.00001129687,0.001958611],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9701721,"threshold_uncertainty_score":0.3656836,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.146145928297451,"score_gpt":0.4845451474897733,"score_spread":0.3383992191923223,"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."}}