Practitioners say AI-generated code should be treated as untrusted
Original title: How organizations can secure their AI-generated code
In SusVibes, agents produced working code up to 57% of the time, but only 11.8% of solutions were secure.
CSO Online reports security leaders treating AI-written code as untrusted input after vibe-coded apps reached production, including one EU case described on Reddit. SusVibes authors say unguided agents often ship working but insecure implementations of real feature requests. Datadog, Mimecast, and Semgrep practitioners recommend pull-request and editor checks, a named human owner, and specialist review for authentication, payments, personal data, and infrastructure.
Unconfirmed: The Reddit account of an EU company shipping a vibe-coded customer-data app is anecdotal and not independently verified.
- SusVibes used 186 feature requests from open-source projects.
- Named assistants include Claude Code, Cursor, GitHub Copilot, and Amazon Q Developer.
- Phantom dependencies and Hugging Face model backdoors are cited as supply-chain risks.
- Organizations
- Carnegie Mellon University
From the original article
8 min read at csoonline.com
The IT manager didn’t know the company hired a “vibe coder” straight out of school. By the time they met a few months later, the new hire had already built a web application that operated with customer data and private contracts. The app ran under a domain the new hire controlled, stored data on Supabase’s free plan, and deployed directly from his GitHub repository to Cloudflare’s serverless infrastructure.
csoonline.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.