{"id":"W7126127877","doi":"","title":"The impact of AI bots and crawlers on open repositories: Results of a COAR survey, April 2025","year":2025,"lang":"","type":"article","venue":"Lincoln (University of Nebraska)","topic":"AI in Service Interactions","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Search engine indexing; Crawling; Webometrics; State (computer science); Web crawler; Service (business)","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.00772344,0.0002515516,0.0003061602,0.004421834,0.001829821,0.002971545,0.0006117026,0.001119025,0.003120315],"category_scores_gemma":[0.02989284,0.000344335,0.0002855288,0.005247006,0.0009053848,0.004025823,0.002129274,0.0008742669,0.002663744],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0014345,"about_ca_system_score_gemma":0.001943909,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0181987,"about_ca_topic_score_gemma":0.02790676,"domain_scores_codex":[0.9928567,0.002743707,0.0006995052,0.0005408507,0.00235161,0.0008074996],"domain_scores_gemma":[0.938095,0.02117755,0.01593822,0.002279251,0.01859366,0.003916312],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0003968424,0.0003976711,0.6910062,0.001145241,0.0001141819,0.0006289615,0.04966839,0.0002244529,0.001847889,0.001303881,0.08670043,0.1665658],"study_design_scores_gemma":[0.00001831077,0.0004361403,0.8078755,0.0004977458,0.000077854,0.0008001361,0.0742621,0.0006376185,0.001245387,0.0001758974,0.1138831,0.00009016282],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9628188,0.003967268,0.0009417001,0.007987324,0.00008018853,0.0001270253,0.004406985,0.0003326135,0.01933792],"genre_scores_gemma":[0.9629171,0.006005289,0.001799124,0.006349491,0.0001936354,0.0002252712,0.003910393,0.0001972118,0.01840248],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9970285,"threshold_uncertainty_score":0.04084593,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02541559873155507,"score_gpt":0.3157427550446062,"score_spread":0.2903271563130511,"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."}}