Simon Willison · 2026-05-06 · notable
Simon Willison: Vibe Coding and Agentic Engineering Are Getting Closer Than I'd Like
Willison confronts the uncomfortable truth that as AI coding agents become more reliable, even careful professionals are trusting their output without reviewing every line — blurring the line between responsible engineering and vibe coding.
As AI agents get more reliable, even careful engineers are skipping code review.
What is it?
Simon Willison's personal reckoning with how his own engineering practice has shifted as AI coding agents have become more reliable — and what that means for professional responsibility.
How does it work?
The essay traces how Willison's trust in AI-generated code has grown to the point where he sometimes deploys without reviewing individual lines, crossing a threshold he previously criticized.
Why does it matter?
The engineering community is grappling with where the line between 'professional AI-assisted development' and 'irresponsible vibe coding' actually lies — Willison is one of the most credible voices to ask it.