Breaking AI: Live coding and hacking applicattions with Generative AI
summary
Live coding uses GitHub Copilot to build a Spring Boot app, then CI and an IDE scanner flag a SQL injection introduced by concatenated user input, a path traversal on file uploads that writes to / (Docker default user often root), and stored XSS from unsanitized product data; Burp Suite confirms exploitation and prepared statements plus path normalization fix the issues. Cites surprising stats: ~97% of devs use AI coding tools, teams 96%, and AWS claims 57% faster.
AI-written summary · about
chapters (9)
- 0:00Introduction to AI in dev
- 2:30GenAI in development risks
- 5:53AI reliability and trust
- 9:25Data privacy and trust center
- 11:52Live demo: SQL injection
- 20:45Remediating SQL injection
- 24:33Live demo: Path traversal
- 34:50Cross-site scripting (XSS)
- 39:27Learn and secure practices
related talks
46mBreaking AI: Live coding and hacking applications with Generative AI
44mBreaking AI: Live Coding and Hacking Applications with Generative AI
30mWhat Are the Security Risks Associated with AI-Generated Code? 🤔
28mAI-Generated Code: Unmasking the Security Pitfalls
58mSecuring Code in the Age of AI
29mHacking With AI and Hacking AI: Two Sides of the Same Problem
30mThe AI Revolution in Coding
30mHuman vs AI: How to Ship Secure Code