Investigating Threats to AI: Incident Response to Compromised AI Workloads

As organizations adopt generative AI on AWS, security teams must extend investigation capabilities into the AI layer. In this hands-on workshop, you trace a compromised credential from initial exposure through privilege escalation into Amazon Bedrock knowledge bases and agents. Using Amazon GuardDuty, AWS Security Hub, AWS CloudTrail, and Amazon Athena, you reconstruct the full event chain and perform investigation and response for threats that move laterally from traditional infrastructure into generative AI workloads.