Article

WitnessAI Inc. has launched Agentic Security, a new capability designed to enhance the oversight and protection of autonomous and semi-autonomous AI agents within enterprises. This solution provides key features such as bidirectional runtime defense against prompt injection and real-time tokenization of sensitive data, addressing the growing need for security as AI agents interact with various systems. CEO Rick Caccia emphasized that Agentic Security offers a unified approach to securing AI interactions, in contrast to existing fragmented methods.