Stop Your Legacy Infrastructure from Hijacking Your AI Agents
Summary: At the Gartner Security & Risk Management Summit, XM Cyber's Zur Ulianitzky described how attackers reach AI agents through legacy infrastructure rather than attacking the AI directly, citing that roughly 71% of organizations are piloting AI agents and 31% have moved them into production workflows. He cited Infosecurity Magazine reporting that 70% of organizations grant AI systems more privileged access than a human in the same role, and that organizations with over-privileged AI reported a 76% incident rate versus 17% for those enforcing least privilege.
What's compelling: A modeled attack path chained a perimeter Tomcat vulnerability through an Active Directory delegation misconfiguration into a developer's cloud credentials and ended at an AI agent's knowledge base, with each layer's finding rated only moderate on its own. With a 76% incident rate for over-privileged AI against 17% under least privilege, the exposures your tools already flag as low priority are the ones that compose into a route to your agents.
Read the original at The Hacker News ↗
- Topics
- AI Security
- Related
- SECURITY AFFAIRS AI-CYBERSECURITY NEWSLETTER ROUND 8 Cyber Risks · Nov 15, 2026Pentagon Renames CDAO ‘Super Intelligence Office’ Cyber Risks · Nov 15, 2026186 Fake AI Domains Redirect Users to Click Fix Lures, Browser Malware, and Search Hijackers Cyber Risks · Nov 12, 2026