{"id":"W4385645253","doi":"10.48550/arxiv.2308.02435","title":"Designing Fiduciary Artificial Intelligence","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Ethics and Social Impacts of AI","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"York University; Microsoft; National Science Foundation","keywords":"Fiduciary; Duty of loyalty; Duty; Context (archaeology); Principal (computer security); Audit; Computer science; Work (physics); Law; Knowledge management; Business; Engineering; Political science; Computer security; Accounting","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.04274802,0.0005352086,0.0008145653,0.00243919,0.004971472,0.008105475,0.002962413,0.003063018,0.002170244],"category_scores_gemma":[0.07609732,0.001190349,0.001054698,0.001145035,0.01949604,0.01098179,0.008212959,0.004595255,0.000798076],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004557978,"about_ca_system_score_gemma":0.008613703,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002592999,"about_ca_topic_score_gemma":0.002391216,"domain_scores_codex":[0.9508236,0.03254098,0.003078743,0.004228067,0.007534887,0.00179374],"domain_scores_gemma":[0.9376718,0.03162653,0.004788894,0.01844499,0.006044815,0.001423088],"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":[0.00002472281,0.00003766698,0.001119526,0.0001007749,0.00001901094,0.00008408146,0.003987664,0.004462549,0.0009681095,0.9604275,0.001016585,0.02775192],"study_design_scores_gemma":[0.00003855306,0.0000889459,0.0004536476,0.0002909831,0.00003253374,0.0002204865,0.00154897,0.03364959,0.005359742,0.8840216,0.07423992,0.00005514567],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02460578,0.0003485281,0.9359562,0.005251077,0.000117233,0.0006240412,0.00003740015,0.0005427237,0.03251707],"genre_scores_gemma":[0.3184269,0.0003506158,0.6743866,0.001028677,0.00006162851,0.0007407682,0.00007460621,0.0001709536,0.004759279],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.04274802,"threshold_uncertainty_score":0.2260758,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3967637717273483,"score_gpt":0.3047609096039285,"score_spread":0.09200286212341985,"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."}}