Vibe Coding vs Corporate Governance: AI Innovation Without Breaking Compliance
summary
Gartner warns four in ten organizations may be forced to shut down AI agents by 2027 if governance is uniform; empirical studies find ~40% of AI‑generated code contains vulnerabilities. Proposes a three‑pillar solution: sandbox paradigm with mock APIs to keep production data safe, CI/CD static analysis and secret scanning that blocks hardcoded credentials, and compliance as code (Open Policy Agent) that intercepts PII leaks. Describes an automated feedback loop to LLM so pipelines fail fast, fix, and re-run — minutes to usable proof‑of‑concepts, humans only for high‑risk releases.
AI-written summary · about
chapters (16)
- 0:00Introduction
- 1:03Gartner’s AI governance warning
- 2:28Slow IT processes frustration
- 4:35Risks of ungoverned AI code
- 8:08Automated governance framework
- 9:35Sandbox for safe AI experiments
- 11:23Shift security left with automation
- 13:36Compliance as code enforcement
- 16:04GDPR breach prevention demo
- 19:36Key takeaways
- 21:21Closing: Business-IT compliance gap
- 23:06Q&A: Empowerment vs governance
- 24:55Q&A: Governance for AI innovation
- 26:20Q&A: Mandatory AI code scanning
- 27:23Q&A: Human review for AI apps
- 28:49Q&A: Managing AI code vulnerabilities
related talks
24mAI Governance for developers: A practical framework for deploying autonomous coding agents safely…
38mVibe-Coding Your Way into a Security Nightmare
08mWhy AI Governance Matters More Than Ever
24m2025: year of Ai dilemma - ethics, regulations and innovations
42mCode Without Caution: AI's Path to the Next Digital Disaster
46mPanel Discussion: Beyond vibe coding: Expanding AI’s role in delivery
29mVibecoding Promise and Peril- 24mAI Governance for developers: A practical framework for deploying autonomous coding agents safely…