{"id":"W4386580576","doi":"10.1007/s44206-023-00063-1","title":"Lessons Learned from Assessing Trustworthy AI in Practice","year":2023,"lang":"en","type":"article","venue":"Digital Society","topic":"Ethics and Social Impacts of AI","field":"Social Sciences","cited_by":28,"is_retracted":false,"has_abstract":true,"ca_institutions":"Cégep André Laurendeau; Université du Québec à Montréal","funders":"Horizon 2020 Framework Programme; Connecting Europe Facility; National Institutes of Health; European Commission; Foundation for the National Institutes of Health","keywords":"Trustworthiness; Computer science; Psychology; Data 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":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.001295449,0.0001016,0.0001613032,0.00002541991,0.0006453678,0.001747682,0.0002230503,0.0002668987,0.00002625136],"category_scores_gemma":[0.00341467,0.0001094903,0.0001776575,0.0008644799,0.0002954726,0.003358075,0.00009328475,0.0005163335,0.0001485401],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001662144,"about_ca_system_score_gemma":0.0004393642,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005492439,"about_ca_topic_score_gemma":0.0009338772,"domain_scores_codex":[0.9984313,0.0001227974,0.0001763538,0.0002583839,0.000540361,0.0004708337],"domain_scores_gemma":[0.9982913,0.001176237,0.00008604105,0.0001311985,0.0001672901,0.0001479134],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00002502066,0.0004571623,0.0281355,0.00001939724,0.0002569357,0.00009658209,0.4252962,0.0001179794,0.0002658677,0.1476628,0.07311386,0.3245526],"study_design_scores_gemma":[0.0004722454,0.00002075714,0.02679562,0.00005604667,0.00002296207,1.743098e-7,0.2175597,0.0003162187,0.00001387464,0.1493355,0.605009,0.000397928],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4316932,0.00008911214,0.0000712816,0.2551278,0.000448244,0.0001712796,0.00005371818,0.0003957308,0.3119496],"genre_scores_gemma":[0.993207,0.000616308,0.0001823582,0.003899699,0.0003460657,0.000005422838,0.00004853547,0.0000181733,0.001676407],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5615138,"threshold_uncertainty_score":0.9992886,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1539208411819251,"score_gpt":0.4696216618800328,"score_spread":0.3157008206981077,"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."}}