Who’s Going to Secure the Code Our Army of Robots Are Going to Be Writing?
summary
Argues that humanity needs AI security engineers because LLMs push developers to produce 25% more code today and potentially 100–500% more soon while security teams are already outnumbered roughly 100:1. Shows concrete tooling: runtime protections like RASP, static checks with Semgrep and libCST, and an open-source code-mod framework called Cod modder that orchestrates find-and-replace rules and auto-remediations; demos and a released repo aim to prove the approach can scale and actually fix findings.
AI-written summary · about
chapters (10)
- 0:00Securing AI-generated code
- 4:24Security risks in LLM code
- 8:41Why AI struggles with secure code
- 12:22Secure development lifecycle
- 16:20Scaling security with code output
- 18:09Making insecure code hard to write
- 21:18Protecting against code exploits
- 25:08Automating security tool triage
- 27:38Codemodder for automated fixes
- 30:07Future outlook and Q&A
related talks
19mIf AI Writes the Code, Who Runs Production?
30mHuman vs AI: How to Ship Secure Code
58mSecuring Code in the Age of AI
41mCode Security Reinvented Navigating the era of AI
33mShipping Fast, Breaking Silently: Why AI-Generated Code Needs a Second Opinion
26mSmarter Code, Sneakier Risks: Supply Chain Security in the Age of AI
59mUsing AI for Secure Code Creation: Enhancing Software Security
01h 24mOffensive Cybersecurity by AI Promises and Pitfalls