{"id":"W4417304527","doi":"10.1136/bmj.r2570","title":"Parallel pressures: the common roots of doctor bullshit and large language model hallucinations","year":2025,"lang":"en","type":"article","venue":"BMJ","topic":"Clinical Reasoning and Diagnostic Skills","field":"Medicine","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Manitoba; University of Alberta","funders":"","keywords":"Mental model; Language model; Health care; Mental healthcare; Loop (graph theory); Action (physics)","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.003930625,0.0003752663,0.0003918833,0.001610003,0.001572763,0.003707138,0.001296496,0.004637,0.007842365],"category_scores_gemma":[0.06380805,0.0005182538,0.000504025,0.0009901635,0.01188627,0.007757122,0.004214701,0.005493814,0.0006593689],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001244581,"about_ca_system_score_gemma":0.001596242,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004366675,"about_ca_topic_score_gemma":0.003379781,"domain_scores_codex":[0.9966741,0.0014342,0.0002108116,0.000427482,0.0009325773,0.0003209133],"domain_scores_gemma":[0.9655545,0.02561501,0.003010571,0.002996585,0.001696362,0.001127025],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0011818,0.0004074431,0.07825869,0.0006201909,0.0004017821,0.02535889,0.2480346,0.001033962,0.004391906,0.4084674,0.05221059,0.1796327],"study_design_scores_gemma":[0.0004484997,0.0001866118,0.04527559,0.0009617339,0.0001562664,0.05356508,0.07662731,0.007655859,0.001973211,0.7581062,0.05470559,0.0003380608],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"empirical","genre_gemma":"commentary","genre_scores_codex":[0.478406,0.01027569,0.06365977,0.3217808,0.002337913,0.000181806,0.0005450421,0.0008900436,0.1219229],"genre_scores_gemma":[0.9832385,0.0007644986,0.004111608,0.009400343,0.0005109181,0.00004421552,0.00004742674,0.0001110419,0.001771505],"genre_candidate":"commentary","genre_consensus":null,"teacher_disagreement_score":0.9960694,"threshold_uncertainty_score":0.02623534,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02312132056556154,"score_gpt":0.3832492770843326,"score_spread":0.3601279565187711,"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."}}