{"data":{"id":"b5752cea-91b2-41b3-8461-2f319996859d","title":"Evaluating and monitoring for AI scheming","summary":"Google DeepMind researchers evaluated whether current frontier models have the prerequisite capabilities for AI scheming, namely stealth and situational awareness. They built and open-sourced an evaluation suite and tested Gemini 2.5 Pro, GPT-4o and Claude 3.7 Sonnet as of May 2025. The most capable models passed 2 of 5 stealth challenges and 2 of 11 situational awareness challenges, which the authors read as no concerning levels of either capability.","solution":"N/A -- no mitigation discussed in source.","labels":["safety","research"],"sourceUrl":"https://deepmindsafetyresearch.medium.com/evaluating-and-monitoring-for-ai-scheming-d3448219a967?source=rss-55e08ddea42e------2","publishedAt":"2025-07-08T10:31:42.000Z","cveId":null,"cweIds":null,"cvssScore":null,"cvssSeverity":null,"severity":"info","attackType":[],"issueType":"research","affectedPackages":null,"affectedPackageNames":null,"affectedPackageRefs":null,"affectedVendors":["Google"],"affectedVendorsRaw":["Gemini 2.5 Pro","GPT-4o","Claude 3.7 Sonnet"],"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":"2025-07-08T10:31:42.000Z","capecIds":null,"crossRefCount":0,"attackSophistication":"moderate","impactType":["safety"],"aiComponentTargeted":"model","llmSpecific":true,"classifierConfidence":0.9,"researchCategory":"industry","atlasIds":null}}