SentinelOne to Integrate AI Security with Amazon Bedrock AgentCore — SentinelOne, Inc.

Event summary

SentinelOne will integrate its Prompt Security offering with Amazon Bedrock AgentCore to provide runtime guardrails for AI agents.
The integration will enable real-time, deterministic controls at the gateway for agent traffic, including agent-to-tool, agent-to-LLM, and agent-to-agent communications.
SentinelOne’s detection signals will feed into AgentCore’s policy engine, allowing for automated enforcement of security controls.
The integration is expected to be available later in 2026, with existing Prompt Security licenses applicable through a Bring Your Own License path.

The big picture

This integration underscores the growing importance of securing AI agents as they become foundational to enterprise operations. By extending its security controls to agentic workflows, SentinelOne aims to address the probabilistic nature of AI security while providing deterministic enforcement. The move reflects broader industry trends towards integrating security measures into AI development and deployment pipelines, particularly as enterprises scale their AI initiatives on cloud platforms like AWS.

What we're watching

Market Adoption: How quickly enterprises will adopt this integration to secure their AI workflows on AWS.
Competitive Response: Whether competitors will introduce similar security integrations for AI agents.
Regulatory Compliance: The pace at which regulatory frameworks will evolve to address AI agent security.