Enterprises are racing to put AI agents into production, but security will likely slow them down.
Â
AWS gives teams the building blocks to build agentic applications on a secure foundation. Amazon Bedrock, AgentCore, Guardrails, IAM, KMS, and CloudTrail cover infrastructure, identity, data, and governance from day one.
Â
But once agents are live, security teams face a new question: what happens when an attacker manipulates the model and uses the agent to reach downstream tools and APIs?
Â
That is where Salt comes in. Salt's Agentic Security Platform, now with native AI Detection and Response (AI-DR), connects a prompt injection to the compromised agent, the MCP tool it calls, the backend API behind it, and the business data at risk, so one attack appears as one story instead of five disconnected alerts.
In this session, AWS and Salt show how the two layers work together.