{"id":"W4415230309","doi":"10.1609/aies.v8i3.36776","title":"From Model Multiplicity to Prompt Multiplicity: Emerging Arbitrariness Concerns in the Age of Generative AI","year":2025,"lang":"en","type":"article","venue":"Proceedings of the AAAI/ACM Conference on AI Ethics and Society","topic":"Ethics and Social Impacts of AI","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Arbitrariness; Generative grammar; Phenomenon; Multiplicity (mathematics); Generative model","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":[],"consensus_categories":[],"category_scores_codex":[0.003676549,0.0002147348,0.0004196766,0.00004676103,0.001085026,0.0002949247,0.001271305,0.0005063468,0.000006076279],"category_scores_gemma":[0.004134906,0.0001504527,0.0001970291,0.0005589656,0.001088846,0.0002707551,0.0004536673,0.002127414,4.020657e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008269902,"about_ca_system_score_gemma":0.0007167212,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01099173,"about_ca_topic_score_gemma":0.002340031,"domain_scores_codex":[0.997919,0.0001337118,0.0004105903,0.0003909276,0.0007826881,0.0003630789],"domain_scores_gemma":[0.9971176,0.0009313123,0.0002405311,0.0002376117,0.001389016,0.00008391208],"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":[0.00002642628,0.0001092992,0.00406618,0.0000883174,0.00005507678,1.523302e-7,0.4604012,0.00007359108,0.01048618,0.5229875,0.001370131,0.0003359844],"study_design_scores_gemma":[0.00110566,0.0001139772,0.01377752,0.001042286,0.00009632922,1.022301e-7,0.1604767,0.02810147,0.008938528,0.7846751,0.001161314,0.0005110009],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7852563,0.00003203043,0.0003563158,0.201653,0.0001475416,0.0007872669,0.00004780242,0.00002367781,0.0116961],"genre_scores_gemma":[0.9789161,0.000494549,0.001355524,0.01876381,0.0001032062,0.00003944764,0.000002069782,0.0000101332,0.000315144],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2999245,"threshold_uncertainty_score":0.9955941,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1243715393619598,"score_gpt":0.4205873121364219,"score_spread":0.2962157727744621,"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."}}