{"id":"W4409403219","doi":"10.1007/978-3-031-87766-7_22","title":"Gender and Race Bias in Consumer Product Recommendations by Large Language Models","year":2025,"lang":"en","type":"book-chapter","venue":"Lecture notes on data engineering and communications technologies","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Victoria","funders":"","keywords":"Race (biology); Product (mathematics); Psychology; Computer science; Sociology; Gender studies; Mathematics","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":[0.000526052,0.0001905656,0.0002569089,0.000362678,0.0002903512,0.00008448098,0.0009750783,0.0002352703,0.00001076305],"category_scores_gemma":[0.001159223,0.000180742,0.00002582668,0.0001972903,0.0001923748,0.0001200205,0.0009059577,0.0006234237,0.000001638056],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004210206,"about_ca_system_score_gemma":0.00005959794,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001580369,"about_ca_topic_score_gemma":0.0008491445,"domain_scores_codex":[0.9990972,0.00006017938,0.0002084539,0.0003550122,0.0001208761,0.0001582757],"domain_scores_gemma":[0.9969591,0.001381034,0.00008012963,0.0015105,0.00004751866,0.00002168906],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00000346314,0.00005189249,0.00006319403,0.00005453442,0.0001407264,0.000001135125,0.001355127,0.0007744923,0.00002239591,0.5484731,0.003266013,0.4457939],"study_design_scores_gemma":[0.000192671,0.00001241041,0.00002838543,0.0002996874,0.0001253253,0.000001620995,0.0006915883,0.03646503,0.00003546576,0.06899365,0.8926616,0.0004925131],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"empirical","genre_scores_codex":[0.0004783127,0.4251305,0.3310682,0.1078194,0.0003267518,0.002134141,0.004476891,0.003815734,0.12475],"genre_scores_gemma":[0.5862663,0.1729705,0.2173547,0.0004693274,0.00006914958,0.0001500244,0.006810109,0.00008799753,0.01582186],"genre_candidate":"review","genre_consensus":null,"teacher_disagreement_score":0.8893957,"threshold_uncertainty_score":0.7370445,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09907225529256185,"score_gpt":0.3587585533588921,"score_spread":0.2596862980663302,"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."}}