Radware enhances AI agent security with compliance reporting and endpoint protection

Radware Ltd. has expanded its Agentic AI Protection product with new features aimed at improving security and compliance for enterprises increasingly relying on AI agents. The update includes enhanced visibility into AI agent activity, compliance reporting aligned with major regulatory standards, and new protection for agents running directly on developer machines, as reported by Silicon Angle.

Radware's update provides improved monitoring and mapping of agent interactions and dependencies, allowing security teams to understand how autonomous systems connect to enterprise resources. A key addition is audit-ready reporting designed to meet standards like ISO 42001, the EU AI Act, and the NIST AI Risk Management Framework, offering companies a documented trail for AI risk management. Notably, the product now extends protection to AI agents operating on developer endpoints, such as Claude Code, closing a security gap for tools not hosted in SaaS environments. This allows for monitoring agent behavior, governing tool usage, and safeguarding sensitive data on local machines, covering both SaaS and developer-hosted agents.

Source: Silicon Angle