{"data":{"id":"cf2120e5-1c87-46b5-8450-c41f969375fe","title":"Microsoft Researchers Show AI Agents Can Be Backdoored by Editing Fewer Than 10 Neurons","summary":"Microsoft researchers showed that an AI model can be backdoored by editing fewer than 10 rarely used neurons, without retraining. When a chosen trigger appears, the model outputs an attacker's command, which an agent using the model would run with the agent's own permissions, so the damage depends on what the agent is allowed to do. The attacker must be able to change the model's weights, and the researchers call the method Lazarus.","solution":"N/A -- no mitigation discussed in source.","labels":["security","research"],"sourceUrl":"https://thehackernews.com/2026/11/microsoft-researchers-show-ai-agents.html","publishedAt":"2026-11-11T07:21:05.000Z","cveId":null,"cweIds":null,"cvssScore":null,"cvssSeverity":null,"severity":"info","attackType":["other"],"issueType":"news","affectedPackages":null,"affectedPackageNames":null,"affectedPackageRefs":null,"affectedVendors":["Microsoft"],"affectedVendorsRaw":["Microsoft","Nemotron-Terminal-8B","Qwen3.6-27B","Gemma-4-31B-IT","Claude Code","Opus 4.5"],"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":"2026-11-11T07:21:05.000Z","fixReleasedAt":null,"daysToFix":null,"fixCheckedAt":null,"capecIds":null,"crossRefCount":0,"attackSophistication":"advanced","impactType":["integrity","safety"],"aiComponentTargeted":"model","llmSpecific":true,"classifierConfidence":0.85,"researchCategory":null,"atlasIds":null}}