TL;DR: AI-generated code can work functionally while still carrying unresolved authentication, authorization and secrets risk, according to Pluto Security. Application security now has to account for review gates, provenance tracking and runtime monitoring before those gaps reach production.
Editorial analysis by NHI Mgmt Group, based on content published by Pluto Security: “Vibe Coding Security: When AI-Driven Development Escapes Security Visibility”.
Key questions
Q: Why does vibe coding increase application security risk?
A: Because it increases the volume and speed of code changes faster than traditional review and scanning can absorb.
Q: What are the signs that AI-assisted development is being used without adequate security controls?
A: Common warning signs include high volumes of generated code, limited human review, security checks arriving late in the SDLC, and developers shipping faster because security feels optional.
Q: What breaks when AI-generated code is reviewed without security gates?
A: What breaks is the assumption that a clean-looking diff is a safe diff.
Practitioner guidance
- Inventory every AI development workflow Map which teams, tools and pipelines are generating code so you can apply the right review and scanning controls to the right systems.
- Tag AI-generated code and commits Label repositories, pull requests and build artefacts that include generated code so reviewers can trace provenance and separate human-authored logic from machine-authored logic.
- Add mandatory security review to generated changes Require peer review and security review for AI-generated code before merge, with special attention to authentication, authorization, logging and error handling.
Bottom line: AI-assisted development can create software that works functionally while still carrying unresolved authentication, authorization and secrets risk.
Glossary terms: Vibe Coding, AI-generated code, Code provenance, Security Visibility
Explore further
View Full Forum → | NHI Foundation Course → | Our Services → | Read the full analysis →
Vibe coding creates governance debt, not just engineering debt: the core issue is that speed-intensive AI development shifts security decisions into workflows that are often under-reviewed and poorly attributed. When teams cannot explain who approved a generated control path, they also cannot prove that the path is trustworthy. That makes software governance a first-class security problem, not a follow-on review task.
A few things that frame the scale:
- 96% of security operations teams report critical blind spots, most commonly in cloud infrastructure (74%) and identity and access behaviour (67%).
A question worth separating out:
Q: How should organisations govern AI tools that analyse source code?
A: Set evidence thresholds, require secondary validation for high-impact findings, and define who owns the final decision. Governance should cover prompt scope, output format, reviewer sign-off, and auditability. That keeps AI in an advisory role and prevents noisy model output from becoming an unreviewed security decision.
👉 Read our full editorial: Vibe coding expands delivery speed while hiding security blind spots