{"id":"W4386330255","doi":"10.31235/osf.io/bq43y","title":"Machine Learning and the Reproduction of Inequality","year":2023,"lang":"en","type":"preprint","venue":"","topic":"Ethics and Social Impacts of AI","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Inequality; Race (biology); Spurious relationship; Key (lock); Reproduction; Computer science; Machine learning; Sociology; Social inequality; Data science; Artificial intelligence; Mathematics; Gender studies; Computer security","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":["sts"],"consensus_categories":[],"category_scores_codex":[0.01498042,0.0004768914,0.0006524125,0.001420213,0.002290137,0.004259863,0.001155236,0.002282813,0.006464476],"category_scores_gemma":[0.07260206,0.0003586952,0.0006438948,0.00177409,0.01660466,0.005569106,0.005301176,0.004116225,0.0008719749],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00231251,"about_ca_system_score_gemma":0.001518623,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002298794,"about_ca_topic_score_gemma":0.001831973,"domain_scores_codex":[0.9852797,0.01161933,0.00024826,0.0009706102,0.001453684,0.0004283595],"domain_scores_gemma":[0.9105905,0.07859011,0.001722868,0.007006231,0.001589062,0.0005011508],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00004241364,0.0000284642,0.003578625,0.00008270456,0.00003813296,0.0001551983,0.002351393,0.007955923,0.0003692757,0.9459708,0.003332743,0.03609451],"study_design_scores_gemma":[0.00001329532,0.00001297199,0.0008854801,0.00005882587,0.000006403022,0.00006785424,0.0002466846,0.01948647,0.0004550658,0.9684644,0.0102906,0.00001197583],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.12818,0.004661737,0.6163789,0.08820488,0.0007638855,0.0001060283,0.0005740903,0.0009804089,0.1601501],"genre_scores_gemma":[0.9329851,0.0008872007,0.05177358,0.002939451,0.0005125817,0.0002480367,0.0001534925,0.0003324607,0.01016815],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9977099,"threshold_uncertainty_score":0.079225,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1451995030291101,"score_gpt":0.4359792346281269,"score_spread":0.2907797315990168,"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."}}