Pattern · Discipline

Defensive UX

Yan · MS / Google / Apple guidelines

Definition. UX design that anticipates ML/LLM imperfection: sets right expectations, makes mistakes recoverable, attributes sources, anchors on familiar patterns, collects feedback.

Five principles

  1. Set right expectations — disclaimers on AI output; surface known limitations on the landing page.
  2. Enable efficient dismissal — Copilot: keep typing to ignore. Chatbots: trivially closeable.
  3. Provide attribution — citations (BingChat); inline quotes; community-recommended badges.
  4. Anchor on familiarity — resist exotic UI. Chat is flexible but high-effort.
  5. Collect feedback in-flow — thumbs / regenerate; variations / upscale; accept-modify-ignore (implicit).

Mapping in the agentic-coding context

Yan drawing on Microsoft's 18 Guidelines for Human-AI Interaction, Google's People + AI Guidebook, Apple HIG for ML

Engineering clients under-invest here because their users are other engineers — "they'll figure it out." They won't; they'll just stop using the agent. 1–2 week deliverable that lifts adoption more than another month of prompt tuning.