{"id":"W4415398751","doi":"10.1109/isi65680.2025.11201116","title":"Chain-of-Ethics: Defending Against Narrative Camouflage Attacks in LLM Moral Judgment","year":2025,"lang":"","type":"article","venue":"","topic":"Law in Society and Culture","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Narrative; Adversarial system; Camouflage; Accountability; Trustworthiness; Vulnerability (computing); Moral reasoning; Normative ethics","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":[0.02223374,0.001192825,0.0007093175,0.001810863,0.002231648,0.006223088,0.002460707,0.00332042,0.003877158],"category_scores_gemma":[0.1417532,0.0008391292,0.001521214,0.0009024063,0.008632952,0.0114769,0.01079863,0.006326871,0.001205421],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002289446,"about_ca_system_score_gemma":0.003362097,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0035631,"about_ca_topic_score_gemma":0.004653345,"domain_scores_codex":[0.9637001,0.027207,0.001301542,0.002768866,0.004357514,0.0006651108],"domain_scores_gemma":[0.8995439,0.06714542,0.006796138,0.020921,0.004349885,0.001243631],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007169124,0.000319875,0.03138774,0.00107708,0.0003148024,0.001070379,0.02569114,0.08896995,0.01045284,0.5412374,0.01949694,0.279265],"study_design_scores_gemma":[0.00006102327,0.0001025703,0.001401319,0.0002575952,0.00006389229,0.0004152924,0.001970363,0.355665,0.008673719,0.6029659,0.02832822,0.00009517408],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1083572,0.0007760496,0.8609045,0.00756396,0.0002342369,0.0003835354,0.0004710423,0.002972601,0.01833694],"genre_scores_gemma":[0.7486814,0.000215557,0.2463983,0.001199487,0.00008491634,0.0002098231,0.0006518395,0.0004538435,0.002104867],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02223374,"threshold_uncertainty_score":0.1175846,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04408818784319817,"score_gpt":0.3513454602048469,"score_spread":0.3072572723616487,"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."}}