{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.002907369,0.0003771003,0.0005476872,0.0004339205,0.001207679,0.0004228135,0.001987189,0.0003669314,0.000360922],"category_scores_gemma":[0.006491035,0.0003317085,0.0002161241,0.003295058,0.00243558,0.0004583293,0.000333644,0.0007818702,0.000145268],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003756168,"about_ca_system_score_gemma":0.0008536063,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002679192,"about_ca_topic_score_gemma":0.005644334,"domain_scores_codex":[0.9958712,0.0002110484,0.001377914,0.000724546,0.0009546251,0.0008606882],"domain_scores_gemma":[0.9971359,0.0005274821,0.000690612,0.0002527845,0.00124443,0.0001487334],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00006880405,0.0002443893,0.004478671,0.00004000012,0.00002079968,7.822638e-7,0.01475123,0.0001103247,0.01623688,0.9337521,0.000610803,0.02968529],"study_design_scores_gemma":[0.00004516021,0.00009944491,0.0005568155,0.0006717434,0.00003363833,4.645609e-7,0.02931175,0.006389812,0.5326216,0.4271772,0.002610885,0.0004814387],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8442463,0.00003218168,0.0003337387,0.01450543,0.001110347,0.0009614893,0.00001682619,0.0001764026,0.1386173],"genre_scores_gemma":[0.9964383,0.00005276927,0.0003814612,0.0008207718,0.0002535176,0.00007260733,0.000001589529,0.0000259163,0.001953013],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5163848,"threshold_uncertainty_score":0.9999135,"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."}}