The Outsized Shadow: Why 5% of AI Users Are Your Biggest Security Risk

**Executive Summary:** Akamai's report warns that a small subset of enterprise AI power users and the proliferation of unmanaged AI tools create significant shadow-AI risks—data leakage, vulnerable browser/IDE extensions (noting ~16.3% with known CVEs), and new attack vectors (vibe hacking, cursorjacking, cometjacking). It highlights that many enterprise AI conversations occur via personal accounts or freemium subscriptions, increasing visibility gaps, and recommends CISOs adopt continuous discovery, contextual DLP, strict SSO enforcement, extension auditing, and treating AI agents as identities.