"Agentjacking" is the latest demonstration of how easily attackers can exploit an AI agent's inability to differentiate between content and instructions.

Tenet Security's Agentjacking exploit hijacks AI coding agents like Claude Code, Cursor, and Codex with an 85% success rate, bypassing all major security

"Agentjacking" is the latest demonstration of how easily attackers can exploit an AI agent's inability to differentiate between content and instructions.