{"id":"W4391307510","doi":"10.1109/smc53992.2023.10394237","title":"How Secure is Code Generated by ChatGPT?","year":2023,"lang":"en","type":"article","venue":"","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":115,"is_retracted":false,"has_abstract":true,"ca_institutions":"Institut National de la Recherche Scientifique; Cégep de l'Outaouais","funders":"","keywords":"Computer science; Code (set theory); Source code; Chatbot; Field (mathematics); Process (computing); Natural language; Code review; Artificial intelligence; Programming language; Computer security; Static program analysis; Software; Software development","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":[0.009800623,0.0006074798,0.0004525116,0.0006681074,0.0008317949,0.001976793,0.001226176,0.002050703,0.003126463],"category_scores_gemma":[0.08448022,0.0004036166,0.0004144085,0.0003609302,0.002360886,0.002710661,0.001586214,0.001496031,0.001810074],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007226338,"about_ca_system_score_gemma":0.00105805,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00097741,"about_ca_topic_score_gemma":0.0008586265,"domain_scores_codex":[0.9894904,0.006205252,0.0004478346,0.001153437,0.002268192,0.0004347579],"domain_scores_gemma":[0.9287267,0.04850535,0.003871848,0.01275739,0.004893662,0.001245073],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.00754573,0.001997394,0.1301825,0.003457175,0.0005991117,0.006218188,0.05511814,0.03481103,0.1837822,0.06139175,0.0494446,0.4654522],"study_design_scores_gemma":[0.0007519005,0.005458789,0.06509575,0.001629055,0.0005348275,0.008255088,0.01549404,0.3861298,0.2300926,0.1337919,0.1521239,0.0006424127],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7711099,0.0007293543,0.1831941,0.007380288,0.0003789358,0.0009816631,0.0009322294,0.01668946,0.01860399],"genre_scores_gemma":[0.9434815,0.0002066219,0.04780995,0.001164965,0.00005387578,0.0003570865,0.0007104737,0.001552429,0.00466303],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.009800623,"threshold_uncertainty_score":0.05183131,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1935670276868286,"score_gpt":0.4295457747303738,"score_spread":0.2359787470435452,"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."}}