Why I Build Guardrails for AI Instead of Trusting It
John Cleary, Fractional CTO
18 November 2026 · 28 min
A fractional CTO explains why he treats AI generated code as something to verify rather than trust, and the guardrails, static analysis and regular pen testing he puts in place instead. He also covers why small teams rely on open source libraries they cannot fully audit, and why single cloud provider risk tends to be accepted rather than solved.
Listen and subscribe
Spotify
Apple Podcasts
YouTube
RSS