Maai & AI
← Stream
weblinks

Why chatbots are not the future of interfaces

Amelia Wattenberger argues that chat interfaces are poorly designed for most use cases because they lack affordances, force users to articulate context that should be built into the interface, and isolate responses without enabling iterative refinement. She advocates for UI patterns that expose parameters and constraints upfront rather than burying them in natural language prompts.

wattenberger.com ↗
Published Tended 17 May 2026 Maturity 🪴 Plant AI 100% Maai Fully written by me.
Mycelium tags, relations, arguments & questions

Relations

  • ChatGPT lacks Affordance
  • Chat interface exhibits Cognitive load
  • Prompt engineering inaccessible-via Affordance
  • Direct manipulation better-fits Chat interface
  • System prompt instance-of Prompt engineering
  • Chat interface counters Iterative refinement

What this post argues beta

  • Blank text inputs create cognitive burden by failing to communicate what a tool can do, how to use it, and what outputs to expect.
  • Context that users manually encode in prompts should instead be surfaced as interface controls, sliders, and configuration options that bake constraints into design.
  • Chat interfaces trap users in sequential isolation, preventing the kind of iterative feedback loops and parameter adjustment that good interactive tools enable.
  • Affordances and constraints are design's responsibility, not user education; interfaces should make correct usage obvious and incorrect usage difficult.

Share