{"id":"W4409362077","doi":"10.1609/aaai.v39i26.34961","title":"Bias Unveiled: Investigating Social Bias in LLM-Generated Code","year":2025,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":16,"is_retracted":false,"has_abstract":true,"ca_institutions":"York University; Concordia University","funders":"","keywords":"Code (set theory); Computer science; Psychology; Programming language","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":[],"consensus_categories":[],"category_scores_codex":[0.009853753,0.0006523724,0.0003845626,0.001494576,0.0009135168,0.001233538,0.001325325,0.001173912,0.001227563],"category_scores_gemma":[0.06617948,0.000246852,0.000523142,0.0008601189,0.0019389,0.00176015,0.002111784,0.001405107,0.0005577597],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001391841,"about_ca_system_score_gemma":0.001794892,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002185435,"about_ca_topic_score_gemma":0.004423357,"domain_scores_codex":[0.9910764,0.005373613,0.0003461086,0.001136936,0.00181172,0.0002551916],"domain_scores_gemma":[0.9420425,0.03942774,0.003414434,0.0108967,0.003551732,0.0006668019],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.002132111,0.001387365,0.1640638,0.001633696,0.0003411967,0.0008658582,0.006251778,0.1726754,0.04058018,0.03639611,0.04629158,0.5273808],"study_design_scores_gemma":[0.0003649637,0.0006054719,0.01886136,0.0001470612,0.00006552893,0.0003716218,0.0008088101,0.8753134,0.04353175,0.03771073,0.02211802,0.0001013156],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8256488,0.00102874,0.1499709,0.001682393,0.0002564072,0.0005332351,0.003110192,0.01006522,0.007704105],"genre_scores_gemma":[0.8974456,0.0001151943,0.09430865,0.0005497724,0.00007044587,0.0004409287,0.004483294,0.0008509099,0.001735212],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.009853753,"threshold_uncertainty_score":0.05211222,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3073948254662761,"score_gpt":0.4053864192582832,"score_spread":0.09799159379200711,"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."}}