Menlo Security targets real-time AI agent security with MARS platform

What happened

Menlo Security introduced Menlo Agent Runtime Security (MARS), a new platform designed to monitor and secure AI agents in real time. As AI agents gain deeper access to enterprise systems and sensitive data, MARS seeks to provide security teams with visibility into the agents’ actions and enforce policies to prevent risks like prompt injection attacks.

Why it matters

AI agents acting autonomously inside enterprise environments create blind spots for security teams. These agents can interact with data, systems, and workflows, potentially exposing organizations to risks that traditional security tools are not built to detect or control. Prompt injection, where malicious inputs compromise AI behavior, is a specific threat that demands real-time monitoring and immediate policy enforcement. MARS targets these gaps by enabling security teams to track AI agents’ runtime behaviors and intervene instantly if suspicious activities arise. This tightens security around AI’s growing role in operations and lowers the risk of data breaches or unauthorized access driven by compromised AI prompts.

What to watch next

The evolution of AI agents will push security vendors to offer more sophisticated runtime controls and visibility tools like MARS. Watch for adoption signals from enterprises integrating AI-driven workflows requiring tighter controls on automated decisions and data handling. Also monitor how MARS handles complex, evolving prompt injection techniques since attackers constantly adapt to new defenses. The success of platforms like MARS will shape how confidently organizations can deploy AI agents at scale without exposing critical assets.

AI Quick Briefs Editorial Desk