{"id":"W4386370663","doi":"10.1145/3616473","title":"Algorithmic Harms in Child Welfare: Uncertainties in Practice, Organization, and Street-level Decision-making","year":2023,"lang":"en","type":"article","venue":"ACM Journal on Responsible Computing","topic":"Ethics and Social Impacts of AI","field":"Social Sciences","cited_by":32,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Agency (philosophy); Welfare; Process (computing); Economic Justice; Computer science; Resource (disambiguation); Public relations; Business; Economics; Knowledge management; Sociology; Political science; Microeconomics; Law","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":["metaresearch","sts"],"consensus_categories":[],"category_scores_codex":[0.006735485,0.0001337287,0.0002262916,0.0007671654,0.001922807,0.000627911,0.0004446183,0.0001710255,0.00002518653],"category_scores_gemma":[0.06659465,0.0001318034,0.00003457088,0.002552472,0.0001302777,0.0006393763,0.0002497225,0.000978631,0.00001228007],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002490681,"about_ca_system_score_gemma":0.0004979889,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000539209,"about_ca_topic_score_gemma":0.001869372,"domain_scores_codex":[0.9974352,0.0007173257,0.0004601403,0.0002302714,0.0006666562,0.0004904197],"domain_scores_gemma":[0.9933997,0.005543614,0.0002689552,0.0001624988,0.0004828999,0.0001423371],"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":[0.0005365387,0.0003520692,0.1537173,0.0000395779,0.0001264643,0.00302979,0.1502105,0.02253116,0.00005248686,0.2009159,0.003011242,0.465477],"study_design_scores_gemma":[0.002992833,0.000374892,0.6444128,0.003757909,0.00003879697,0.0002020806,0.1846347,0.005344191,0.00004133766,0.1192563,0.03782279,0.001121433],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9277883,0.0002354609,0.0003676818,0.06783783,0.000667365,0.0001891427,0.000003002743,0.00007703192,0.002834201],"genre_scores_gemma":[0.9917605,0.001394109,0.005556765,0.0008874375,0.0003328362,4.934296e-7,0.000001224736,0.00002223803,0.00004442011],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4906955,"threshold_uncertainty_score":0.9993765,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04108821450862449,"score_gpt":0.3891098166661087,"score_spread":0.3480216021574842,"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."}}