'Don't Be Nice': Developer argues politeness with AI coding assistants produces worse code
edent · hn · 2026-09-20
A widely discussed blog post argues that being polite with AI coding assistants backfires: models trained with RLHF tend to cave to pushback rather than defend correct solutions, and polite users who accept first drafts ship worse code. The author recommends treating assistants like tools under strict review—explicitly flagging errors, demanding alternatives, and pressure-testing outputs. HN comments debate whether sycophancy is a structural flaw of preference-tuned models.
More from coding & agent
- 70 hands-on cybersecurity projects with full source code — tom_doerr · 2026-09-21
- Dev compares coding models building a coop game: V4.1 outshines Astra's 'pathetic' default taste — teortaxesTex · 2026-09-21
- 'Just 3 lines of code' backfires: dev argues tools should expose complexity, not hide it — willcb · 2026-09-21
- TypeSafe's Jev returns typed decisions with probabilities, not text — here's where it fits in agent loops — prakersh · 2026-09-21
- Jev Engineering gives agents a decision brain, 193x faster and 444x cheaper in tests — agihouse_org · 2026-09-21
- Dev Argues PAW Shouldn't Hide Its Complexity, Points to DSPy as the Better Playbook — willcb · 2026-09-21