Shadow AI and the permissions problem: what to check before handing AI the keys

Security executives from WatchGuard Technologies and Syntax discuss the risks of AI agents being granted access to accounts, files, and permissions without proper oversight, tying the discussion to Cybersecurity Awareness Month. They warn that AI accounts holding user privilege become high-value targets for attackers, and that 'shadow AI' - unauthorized chatbots, plug-ins, extensions, and AI features silently added to existing tools - creates blind spots organizations fail to monitor. They recommend treating AI accounts like bank accounts (MFA, skepticism of links), determining what an AI tool can see, do, and who is accountable before granting access, and prioritizing visibility, then policy, then enforcement. One cited WatchGuard statistic: 95% of malware arrives over encrypted TLS traffic, yet only 20% of deployed devices inspect that traffic.

Nguồn: https://www.itsecurityguru.org/2026/10/02/shadow-ai-and-the-permissions-problem-what-to-check-before-handing-ai-the-keys. 8 Sync News chỉ tóm tắt và dẫn link; bản quyền nội dung thuộc tác giả và nguồn gốc.