Using generative AI to create more secure applications
summary
Generative AI is used to find and fix real issues: Amazon Q Developer scans a TypeScript ECS app, reporting 11 problems (10 in the app) including a hardcoded DB password, permissive 4MB payload, SQL injection and missing range checks for a 1–5 rating. Q suggests Amazon Linux or Alpine images, rewrites the INSERT to an ON DUPLICATE KEY UPDATE upsert, injects secrets via Secrets Manager, and auto-generates a README via a /dev agent.
AI-written summary · about
chapters (10)
- 0:00Introduction to Amazon Q
- 4:57Security engineer scenario
- 5:48Gathering internal knowledge
- 7:30Navigating AWS documentation
- 9:04Remediating infrastructure
- 10:36Code security scanning
- 13:41Implementing code fixes
- 16:02Automating documentation
- 17:55Advanced agent capabilities
- 19:42Conclusion
related talks
44mGenerative AI Security — A Practical Guide to Securing Your AI Application
31mHacking AI Applications
46mBreaking AI: Live coding and hacking applicattions with Generative AI
46mBreaking AI: Live coding and hacking applications with Generative AI
27mSafeguarding Generative AI Applications: Navigating Security Challenges and Best Pract...
59mUsing AI for Secure Code Creation: Enhancing Software Security
40mSecuring Generative AI: Is it all an Illusion?
30mSoftware in the age of human and AI collaboration