{"id":"W4412995430","doi":"10.1080/09540962.2025.2541304","title":"Debate: Evidence-based AI risk assessment for public policy","year":2025,"lang":"en","type":"article","venue":"Public Money & Management","topic":"Ethics and Social Impacts of AI","field":"Social Sciences","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Business; Risk assessment; Risk analysis (engineering); Computer science; 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":"codex-gemma-dda1882f352a","candidate_categories":["sts","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.006012081,0.0002207237,0.0002800483,0.0006409253,0.002152993,0.002400477,0.000956069,0.0001991968,0.0001639417],"category_scores_gemma":[0.00473658,0.00022064,0.0002340644,0.001438981,0.0003322801,0.001527931,0.0002655054,0.0003259973,0.00001808092],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009810902,"about_ca_system_score_gemma":0.002141613,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008151066,"about_ca_topic_score_gemma":0.002634977,"domain_scores_codex":[0.996703,0.0005250248,0.000404161,0.0004951481,0.0008910759,0.0009815353],"domain_scores_gemma":[0.9973283,0.0007367006,0.0002134238,0.0004904317,0.0008457326,0.0003854402],"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":[0.000008656802,0.000186007,0.005430613,0.00007749024,0.0001698342,0.000002335036,0.0005310877,0.000017027,0.000001660272,0.8420163,0.06268155,0.08887746],"study_design_scores_gemma":[0.0007391128,0.00005788857,0.01552439,0.00009330311,0.0001026621,3.350129e-8,0.001741533,0.001244181,0.000006275897,0.1653609,0.8148485,0.0002812818],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"commentary","genre_gemma":"empirical","genre_scores_codex":[0.002098468,0.000315179,0.05473516,0.7567075,0.0007337025,0.001657536,0.00002347319,0.0002638402,0.1834652],"genre_scores_gemma":[0.9600883,0.002234226,0.003583204,0.02445151,0.0005403401,0.0005594281,0.00002099181,0.00002268649,0.008499294],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9579899,"threshold_uncertainty_score":0.999146,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08906558772105834,"score_gpt":0.4265185242125413,"score_spread":0.337452936491483,"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."}}