{"id":"W6949660431","doi":"10.5281/zenodo.15982693","title":"Safe AI Doctrine","year":2025,"lang":"en","type":"preprint","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Ethics and Social Impacts of AI","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Fields Institute for Research in Mathematical Sciences","funders":"","keywords":"Doctrine; Coherence (philosophical gambling strategy); Robustness (evolution); Reinforcement learning; Identity (music); Audit; Revocation; Embedding","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":[0.00922031,0.0010164,0.0008957048,0.0013958,0.003263091,0.008203123,0.003007904,0.004066399,0.01150808],"category_scores_gemma":[0.03076221,0.0006947896,0.001435631,0.0007008294,0.01402958,0.01042782,0.007572671,0.008171825,0.003421396],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003033396,"about_ca_system_score_gemma":0.005796557,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002807845,"about_ca_topic_score_gemma":0.0018389,"domain_scores_codex":[0.9869731,0.004451874,0.001068549,0.002259246,0.004273061,0.0009741668],"domain_scores_gemma":[0.9881031,0.00444679,0.0009108727,0.003478784,0.002359527,0.0007010104],"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":[0.0000052813,0.000007495803,0.00006038067,0.00002208038,0.000005189062,0.00002154829,0.0001339711,0.001466698,0.0001891444,0.9928335,0.001035741,0.004219058],"study_design_scores_gemma":[0.00001083074,0.00001630169,0.00003663305,0.00005178805,0.000004776153,0.00005016856,0.00007692179,0.008001676,0.0004153092,0.9727482,0.01857554,0.00001170845],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"commentary","genre_scores_codex":[0.007881761,0.0007018247,0.8148476,0.01041562,0.0005005432,0.0002355513,0.0002599874,0.0008999377,0.1642572],"genre_scores_gemma":[0.6100242,0.001026595,0.349137,0.0053991,0.0007186712,0.0008561088,0.00058005,0.0007489786,0.03150931],"genre_candidate":"commentary","genre_consensus":null,"teacher_disagreement_score":0.01150808,"threshold_uncertainty_score":0.04876226,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07548280308370091,"score_gpt":0.3663557902109167,"score_spread":0.2908729871272158,"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."}}