This video from AWS re:Inforce 2023 explores how to securely build generative AI applications using Amazon Bedrock. The speakers discuss the architecture, data flows, and security aspects of Amazon Bedrock, which provides easy access to foundation models from Amazon and other AI companies. They explain how Bedrock integrates with AWS security services like KMS, CloudTrail and IAM to protect customer data and control access. Key points include Bedrock's data privacy guarantees, encryption of data in transit and at rest, and options for multi-tenant or single-tenant model deployments. The presenters also cover model fine-tuning, networking configurations, and compare Bedrock to other AWS machine learning offerings. Overall, the session aims to show how Bedrock enables organizations to leverage generative AI capabilities while maintaining security and control over their data and models.