{"id":"W4381929569","doi":"10.1002/aaai.12099","title":"Maximizing AI reliability through anticipatory thinking and model risk audits","year":2023,"lang":"en","type":"article","venue":"AI Magazine","topic":"Ethics and Social Impacts of AI","field":"Social Sciences","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"Saint John Regional Hospital","funders":"","keywords":"Interpretability; Audit; Robustness (evolution); Reliability (semiconductor); Bridge (graph theory); Computer science; Corporate governance; Risk analysis (engineering); Work (physics); Management science; Knowledge management; Process management; Engineering; Business; Artificial intelligence; Accounting","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":[0.02591257,0.0009363136,0.0006027294,0.002057639,0.001450533,0.00574918,0.001954224,0.001548709,0.001820673],"category_scores_gemma":[0.08297151,0.0006477177,0.0008238939,0.001091818,0.006674923,0.007321062,0.006591389,0.003580323,0.0002334085],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003384797,"about_ca_system_score_gemma":0.008000262,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002682863,"about_ca_topic_score_gemma":0.003542187,"domain_scores_codex":[0.9783792,0.01449083,0.0009213162,0.001264905,0.003912639,0.001031141],"domain_scores_gemma":[0.9183812,0.0450873,0.01349801,0.01246473,0.008443773,0.002125003],"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":[0.0001965954,0.0003366244,0.01111691,0.0003171752,0.0001285388,0.0002476856,0.006138425,0.2007875,0.003166901,0.6656485,0.002612615,0.1093026],"study_design_scores_gemma":[0.0000337621,0.0001282263,0.001067244,0.000203765,0.00005024861,0.00007721407,0.001071316,0.3013003,0.002444916,0.6884777,0.005081849,0.00006345372],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08951151,0.000271178,0.8749425,0.008745493,0.00006786314,0.000236222,0.00003973862,0.0005807823,0.02560474],"genre_scores_gemma":[0.845829,0.0001424503,0.1527155,0.0002036735,0.00002559919,0.0001181277,0.00002561704,0.00005067982,0.0008893021],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02591257,"threshold_uncertainty_score":0.1370404,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05910301051109396,"score_gpt":0.3872911114209994,"score_spread":0.3281881009099055,"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."}}