Check Point Integrates AI Guardrails into Amazon Bedrock AgentCore

The integration allows Check Point to deploy AI-native guardrails that operate outside the agent’s reasoning loop. By feeding behavioral threat detection and content inspection signals directly into AgentCore policies, the company enables automated enforcement of security protocols. While the detection of threats remains probabilistic, the enforcement mechanism is deterministic, ensuring that every action taken by an agent is subject to a strict allow-or-deny decision.

Adam Ely, General Manager of AI Security at Check Point, noted that the evolution of AI necessitates a shift in how enterprises manage risk. Permissions alone are no longer sufficient to govern autonomous workflows, he argued. By securing the environment where these agents operate, Check Point intends to move beyond simple access control toward ensuring that agents consistently perform authorized, safe actions within complex digital ecosystems.