{"id":"W4417465597","doi":"10.1093/oxfordhb/9780198940272.013.0025","title":"Evaluating the Social Impact of Generative AI Systems","year":2025,"lang":"en","type":"book-chapter","venue":"Oxford University Press eBooks","topic":"Ethics and Social Impacts of AI","field":"Social Sciences","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Mila - Quebec Artificial Intelligence Institute; Simon Fraser University; Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Generative grammar; Context (archaeology); Software deployment; Trustworthiness; Moderation; Social impact; Variety (cybernetics)","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":["sts"],"consensus_categories":[],"category_scores_codex":[0.01895483,0.0006572007,0.000435003,0.002196123,0.001377934,0.006109193,0.001160515,0.001132125,0.01071223],"category_scores_gemma":[0.07633167,0.000287716,0.0004505416,0.002003324,0.004742735,0.004392313,0.003764453,0.001713668,0.0007436561],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.006129142,"about_ca_system_score_gemma":0.002134683,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003008548,"about_ca_topic_score_gemma":0.002948227,"domain_scores_codex":[0.9618641,0.02853805,0.0005887297,0.0007248385,0.007686414,0.000597949],"domain_scores_gemma":[0.8730757,0.1076641,0.004260426,0.004541054,0.009163433,0.001295257],"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":[0.0007276261,0.001065892,0.04151406,0.00177594,0.0004651556,0.0003196369,0.009445705,0.1126793,0.004144179,0.5189551,0.01334473,0.2955628],"study_design_scores_gemma":[0.0001896656,0.003106554,0.04627468,0.002103698,0.0005751397,0.0001848012,0.01773363,0.2480013,0.01251375,0.5983517,0.07076144,0.0002037417],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5319604,0.002345248,0.08071975,0.007622487,0.0002929026,0.001491937,0.0008026263,0.0003136634,0.3744509],"genre_scores_gemma":[0.9615559,0.0006830788,0.02900258,0.0004085799,0.00005832564,0.0007234204,0.0002801492,0.00009798663,0.007189909],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9986221,"threshold_uncertainty_score":0.1002439,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1247313057308181,"score_gpt":0.4104357781871379,"score_spread":0.2857044724563198,"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."}}