David Khourshid: agents rewrite tests to match UI changes — we need behavioral diff testing
DavidKPiano · x · 2026-10-24
Frontend engineer David Khourshid uses a real Codex Desktop example to show how an app can be fully tested and still feel broken to users.
The chain: ship UI/UX version A, fully tested; users get used to A; an agent changes the UI to version B and rewrites the tests to match, not caring about A; users expect A but get B. His screenshot shows Codex Desktop's former archive icon replaced by Pin, with Archive moved into a menu under an odd shortcut.
His takeaway: traditional tests verify correctness but not changes to user habits — the industry needs a new kind of "behavioral diff" test to guard against agents silently altering established interaction patterns.
More from coding & agent
- The phone is the computer: how device makers and model labs battle over personal agents — sujingshen · 2026-10-24
- Creator uses GPT via MCP to drive Runway SEEDANCE 2.5 for film-like video — aziz4ai · 2026-10-24
- Five web-data vendors, 400 lines of glue: every bug this quarter lived in the glue — cup_cake009 · 2026-10-24
- Voice-controlled auto-advancing slides go viral; Tejas Kumar built the same in 2018 — TejasKumar_ · 2026-10-24
- Solo founder: after 6 months I stopped reviewing my AI agents' code — Al_Grigor · 2026-10-24
- Building an enterprise-wide custom agent platform: which UI/harness to use? — Electronic_Sort_6627 · 2026-10-24