{"data":{"id":"0021fde6-d369-42f4-8999-1708878cacf3","title":"Discovering Jailbreaks via Curriculum of Models","summary":"Researchers introduce Curriculum of Models (CoM), a method that trains a jailbreak-discovery investigator model against a sequence of progressively harder versions of a target model, starting from a version that rarely refuses harmful requests. They evaluate CoM on CBRN, explosives, and illicit drug elicitation tasks against the open-weight targets gemma-4-31b-it and gpt-oss-120b. CoM finds successful jailbreaks more reliably than the PRBO baseline and direct training when trained on a single behavior, though PRBO-trained investigators achieve higher success rates in the multi-behavior setting.","solution":"N/A -- no mitigation discussed in source.","labels":["research","security"],"sourceUrl":"https://arxiv.org/abs/2610.38833v1","publishedAt":"2026-10-28T20:23:32.000Z","cveId":null,"cweIds":null,"cvssScore":null,"cvssSeverity":null,"severity":"info","attackType":["jailbreak"],"issueType":"research","affectedPackages":null,"affectedPackageNames":null,"affectedPackageRefs":null,"affectedVendors":[],"affectedVendorsRaw":["gemma-4-31b-it","gpt-oss-120b"],"classifierModel":"claude-haiku-5-5","classifierPromptVersion":"v4","summaryPromptVersion":"v2","headline":null,"headlinePromptVersion":null,"cvssVector":null,"attackVector":null,"attackComplexity":null,"privilegesRequired":null,"userInteraction":null,"exploitMaturity":null,"epssScore":null,"epssCheckedAt":null,"kevDateAdded":null,"advisoryAliases":null,"affectedPackagesSource":null,"affectedPackagesCheckedAt":null,"patchAvailable":null,"disclosureDate":"2026-10-28T20:23:32.000Z","fixReleasedAt":null,"daysToFix":null,"fixCheckedAt":null,"capecIds":null,"crossRefCount":0,"attackSophistication":"advanced","impactType":["safety"],"aiComponentTargeted":"model","llmSpecific":true,"classifierConfidence":0.95,"researchCategory":"preprint","atlasIds":["AML.T0054"]}}