{"id":"W4393177920","doi":"10.2196/59479","title":"The Opportunities and Risks of Large Language Models in Mental Health","year":2024,"lang":"en","type":"preprint","venue":"JMIR Mental Health","topic":"Mental Health Treatment and Access","field":"Psychology","cited_by":166,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Mental health; Business; Psychology; Psychiatry","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001427817,0.000433845,0.000753027,0.0002574415,0.0005013886,0.00006582207,0.000342667,0.0002230097,0.0001646593],"category_scores_gemma":[0.000001342998,0.0003297035,0.0001241641,0.0001390246,0.0001609114,0.0000711296,0.0009564417,0.000924486,0.0000309931],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007942425,"about_ca_system_score_gemma":0.0003999396,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01628254,"about_ca_topic_score_gemma":0.005132753,"domain_scores_codex":[0.9962609,0.0005637098,0.001129336,0.0006872161,0.0003970335,0.0009617681],"domain_scores_gemma":[0.9983971,0.00008407376,0.0005865723,0.0005641478,0.00001420092,0.0003539729],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"qualitative","study_design_scores_codex":[0.001185026,0.002668826,0.00258207,0.01053339,0.0003753474,0.0001059069,0.275305,0.000004586304,0.000005359076,0.07921844,0.09428625,0.5337298],"study_design_scores_gemma":[0.02333447,0.007314271,0.06281606,0.02028614,0.0001184452,0.000350448,0.716355,0.006278548,0.0001334274,0.0537496,0.1055974,0.003666204],"study_design_candidate":"qualitative","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8456388,0.1045993,0.000002569235,0.03046133,0.002934484,0.005267012,0.003972116,0.0001058028,0.007018533],"genre_scores_gemma":[0.9765311,0.01475117,0.00006309116,0.001973925,0.0001345134,0.0007348717,0.00107006,0.00007022241,0.004670996],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5300636,"threshold_uncertainty_score":0.9999155,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1589390950813089,"score_gpt":0.4964173053370983,"score_spread":0.3374782102557894,"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."}}