Protecting data in generative AI applications with Amazon Bedrock
summary
Amazon Bedrock integrates with Amazon Q Developer, Inspector and Detective and enforces encryption in transit (TLS 1.2+) and at-rest with AWS KMS backed by HSMs so keys are tamper‑protected and replicated across AZs. Foundational models do not consume customer data and providers can only write model artifacts to an AWS‑owned S3 account; custom models and data remain in the customer account and are custom models encrypted with customer KMS keys and access is controlled via IAM, VPC endpoints, CloudTrail and optional VPN/Direct Connect.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 1:05Security challenges
- 2:58Security maturity model
- 5:28Understanding generative AI
- 6:58Risks of generative AI
- 10:30Improving security posture
- 12:09Data protection in Amazon Bedrock
- 16:17Safe app development
- 20:13Conclusion
related talks
55mSecurely build generative AI apps & control data with Amazon Bedrock
40mSecure Your GenAI Applications
20mData is Your Differentiator: Building Secure and Tailored AI Systems
41mAccelerate Generative AI Adoption with Amazon Bedrock
24mBuilding GenAI applications with Amazon Bedrock
22mObservability in AI-Powered Applications with Amazon Bedrock
31mLamRAG: From Data to Constructive Insights using Amazon Bedrock
29mCustomizing genAI applications for your business