Generative AI to identify potential risks in architectural diagrams
summary
Demonstration uses Amazon Bedrock with the Claude model to parse PNG architecture diagrams and pasted CloudFormation JSON, returning two lists: best practices found and six actionable deficiencies from the diagram, then many more findings when given IaC. Model identifies use of Secrets Manager for RDS, flags ALB listening on HTTP instead of HTTPS, and runs with Temperature=0 for repeatable outputs. Works as an accelerator but requires human validation.
AI-written summary · about
chapters (5)
- 0:00Introduction to frameworks
- 2:20Defining the scenario
- 3:20Setting up Amazon Bedrock
- 5:42Analyzing architecture diagrams
- 7:30Leveraging infrastructure code
related talks
25mFrom Requirements to Reality: How Solutions Architects Think, Build & Decide
26mGenerative AI driven sensitive data protection
31mAccelerate AWS Well-Architected reviews with Generative AI
52mA Proven Approach on Automated Security Architectural Pattern Validation
47mAutomating Architectural Risk Analysis with the Open Threat Model format
33mEnterprise Architecture Governance in the AI Era
33mMaster enterprise-scale generative AI
50mUsing IAC to Accelerate Threat Modeling