{"id":"W4391247296","doi":"10.2139/ssrn.4699451","title":"Edge of Tomorrow: Evaluating Misinformation and Bias in LLM-Powered Chatbots on Climate Change and Mental Health","year":2024,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Misinformation and Its Impacts","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto; University of Ottawa; Dalhousie University","funders":"","keywords":"Misinformation; Mental health; Climate change; Enhanced Data Rates for GSM Evolution; Psychology; Computer science; Telecommunications; Computer security; Psychiatry; Oceanography; Geology","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":[0.02791569,0.0006403127,0.0008365116,0.001902035,0.002069249,0.002800088,0.001040322,0.003268905,0.01030507],"category_scores_gemma":[0.2873252,0.0004510006,0.0008116327,0.001574311,0.001247816,0.005959193,0.003912723,0.002185306,0.001497449],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002131051,"about_ca_system_score_gemma":0.00177725,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005774443,"about_ca_topic_score_gemma":0.007069945,"domain_scores_codex":[0.9824075,0.01384304,0.0007881671,0.0007661112,0.001746668,0.0004487109],"domain_scores_gemma":[0.4760896,0.4914089,0.01533899,0.006690513,0.006987859,0.003484],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.03895086,0.01282985,0.6124166,0.004753312,0.002823758,0.0006019386,0.04566721,0.01322259,0.002925636,0.009653905,0.01710846,0.2390459],"study_design_scores_gemma":[0.005301318,0.02858128,0.7571971,0.002247419,0.00528949,0.0004001654,0.03100109,0.1181619,0.008043353,0.03025396,0.01290306,0.0006197357],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.989011,0.0002593145,0.002342063,0.0008620245,0.0001248556,0.00045647,0.0007852592,0.0001876988,0.005971292],"genre_scores_gemma":[0.9914063,0.00007163339,0.00463234,0.0004670554,0.00008898865,0.00137332,0.0005167845,0.0000365286,0.001407165],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9720843,"threshold_uncertainty_score":0.147634,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1023672522559795,"score_gpt":0.399468375773169,"score_spread":0.2971011235171894,"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."}}