{"id":"W4401415801","doi":"10.1109/access.2024.3440647","title":"A Systematic Literature Review on AI Safety: Identifying Trends, Challenges, and Future Directions","year":2024,"lang":"en","type":"article","venue":"IEEE Access","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":30,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université du Québec à Chicoutimi","funders":"Mitacs","keywords":"Interpretability; Computer science; Software deployment; Notice; Autonomy; Risk analysis (engineering); Robustness (evolution); Adversarial system; Trustworthiness; Artificial intelligence; Data science; Computer security; Software engineering; Medicine","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0005957484,0.0002194458,0.0003728517,0.0003181342,0.0002072121,0.001114493,0.0008575514,0.00009476313,0.00001386465],"category_scores_gemma":[0.00007283065,0.0001686237,0.0001001445,0.001087797,0.00001846601,0.001651597,0.0002164914,0.000577234,0.00002758929],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005805861,"about_ca_system_score_gemma":0.00002608855,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003629981,"about_ca_topic_score_gemma":0.000009123307,"domain_scores_codex":[0.9982701,0.0002740527,0.0003279179,0.0005817879,0.0003241887,0.0002219871],"domain_scores_gemma":[0.9989555,0.0002208409,0.00009178158,0.0005929724,0.00005762055,0.00008132017],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"systematic_review","study_design_scores_codex":[0.000007344608,0.00004477716,0.000005001501,0.2345254,0.000220224,0.0005471721,0.004885393,0.0002319191,0.0000120347,0.1033302,0.005725445,0.6504651],"study_design_scores_gemma":[0.0005207374,0.0001265323,0.0009121902,0.6749012,0.0004172365,0.001074121,0.0001209523,0.05688234,0.0000419852,0.004246386,0.2594198,0.001336447],"study_design_candidate":"systematic_review","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.00002076435,0.9219408,0.05264308,0.01742035,0.004437631,0.0003099447,0.000003715007,0.000583645,0.002640099],"genre_scores_gemma":[0.02642665,0.9662521,0.001610647,0.002069328,0.002213373,0.0001530005,0.000009550738,0.00005693373,0.00120838],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.6491287,"threshold_uncertainty_score":0.9999225,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02978114980185721,"score_gpt":0.339898914322889,"score_spread":0.3101177645210318,"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."}}