{"id":"W4412813486","doi":"10.20944/preprints202507.2350.v1","title":"A Federated Domain-Specific Architecture for Safe and Scalable Artificial Intelligence","year":2025,"lang":"en","type":"preprint","venue":"Preprints.org","topic":"Scientific Computing and Data Management","field":"Decision Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Scalability; Computer science; Architecture; Domain (mathematical analysis); Computer architecture; Artificial intelligence; Distributed computing; Operating system; Geography","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":["metaepi_narrow","scholarly_communication","open_science","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.00915545,0.0004715173,0.0007267704,0.0007694616,0.000675666,0.001146837,0.002275293,0.000304726,0.0008464017],"category_scores_gemma":[0.00225029,0.0004051039,0.0002878856,0.000945795,0.0003153638,0.0001128175,0.008273744,0.0007527294,0.0009413754],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001057357,"about_ca_system_score_gemma":0.0002329269,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008071565,"about_ca_topic_score_gemma":0.00009124674,"domain_scores_codex":[0.9928789,0.0003594647,0.001475709,0.003390466,0.001285656,0.0006098061],"domain_scores_gemma":[0.9942934,0.001655076,0.0005289718,0.002701191,0.000596204,0.0002251516],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0006949123,0.0005316096,0.01967997,0.0003937199,0.0002884576,0.00002913354,0.00331266,0.01827079,0.001631795,0.05992503,0.0251473,0.8700946],"study_design_scores_gemma":[0.00014837,0.00002294971,0.008077192,0.0002300701,0.00003317026,0.000004487534,0.0007787385,0.007815174,0.004229589,0.5582104,0.4199042,0.0005456079],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2777335,0.0003206629,0.6905688,0.003809116,0.004451721,0.001970939,0.0003314966,0.0002261292,0.0205876],"genre_scores_gemma":[0.9540244,0.0001040968,0.0270732,0.0004516202,0.0005129859,0.000320209,0.0001783081,0.00003668465,0.01729852],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.869549,"threshold_uncertainty_score":0.9998901,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3392369762202782,"score_gpt":0.4346278414541759,"score_spread":0.09539086523389778,"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."}}