AI Security Threats
Understanding the novel security threats introduced by AI agents and the MCP protocol - from malware-laden STDIO servers to data exfiltration via the Lethal Trifecta.
AI agents introduce fundamentally new threat vectors that traditional security tools were not designed to handle. These pages explain the threats Edison Watch defends against.
Topics
MCP Malware Risks
Why STDIO MCP servers are the new shadow IT - unmanaged HTTP clients carrying plaintext credentials with zero enterprise visibility.
The Lethal Trifecta
The three-capability combination that enables AI-driven data exfiltration, and how Edison Watch enforces against it.
AI-Compatible RBAC
Why traditional access control breaks when AI agents act on behalf of users, and how to enforce data boundaries.