{"id":"W4411012657","doi":"10.70777/si.v2i2.14755","title":"The First International AI Safety Report","year":2025,"lang":"en","type":"article","venue":"SuperIntelligence - Robotics - Safety & Alignment","topic":"Ethics and Social Impacts of AI","field":"Social Sciences","cited_by":27,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"Mozilla Foundation; Royal Society; DeepMind","keywords":"Forensic engineering; Engineering","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":["sts"],"consensus_categories":[],"category_scores_codex":[0.002975593,0.0002307411,0.0002483101,0.00006971537,0.003337622,0.0005622286,0.001171456,0.0002358385,0.0003117075],"category_scores_gemma":[0.001470247,0.0001943494,0.0002244197,0.0004054525,0.0009558807,0.0003383554,0.0003171214,0.0004655336,0.00009485552],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008598411,"about_ca_system_score_gemma":0.0007603238,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002228525,"about_ca_topic_score_gemma":0.006232953,"domain_scores_codex":[0.996843,0.000158806,0.00079642,0.0004397736,0.001076006,0.000686031],"domain_scores_gemma":[0.9978321,0.0007078899,0.000141704,0.0005664433,0.0005487502,0.0002031679],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0000716083,0.00011714,0.004081993,0.00001108729,0.0001901844,0.00005158601,0.007026908,0.007174189,0.0000137854,0.966854,0.004090483,0.01031703],"study_design_scores_gemma":[0.0001037077,0.00002837669,0.001417399,0.00007286249,0.0000289141,0.000003501274,0.007287334,0.0003521934,0.0001924271,0.04420989,0.9460683,0.0002351233],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"commentary","genre_gemma":"empirical","genre_scores_codex":[0.0005412044,0.0005409989,0.07639238,0.5307176,0.008911761,0.0009013967,0.00002936283,0.0001943018,0.381771],"genre_scores_gemma":[0.9360189,0.01624316,0.001477622,0.007497282,0.0008339079,0.00004147153,0.00003436574,0.00003597739,0.03781733],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9419778,"threshold_uncertainty_score":0.9979599,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02293576748403459,"score_gpt":0.361102392748586,"score_spread":0.3381666252645514,"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."}}