Human Computer Interaction: The UX of AI
summary
Argues that conversational prompts fail because humans have goals not questions and shows concrete alternatives: a fully hidden prompts UI that watches code and flags a race condition instead of waiting for a query; a questionnaire-driven generator that builds a multi-step chain-of-thought prompt to pick Canadian cities or date-night options; and a verification step that checks claimed business names/addresses with the Google Places API or URL 404 checks to reduce hallucinations and preserve trust.
AI-written summary · about
chapters (6)
- 0:00Introduction and setup
- 2:13History of user interfaces
- 10:03Limitations of current AI interaction
- 19:38Science fiction and AI expectations
- 21:04Building better AI tools today
- 26:33Call to action and conclusion
related talks
55mDon’t Make Me Prompt: Intelligent Interfaces Users Actually Like
57mDesigning Human-AI Experience
37mFinding the Problem to Solve: Human-centered AI
25mArtificial intelligence & design: imagining the future of UX
42mUX+AI: Mastering the UX of AI-Powered Products
27mMisplAIced: Good UI Needs Good UX
03mDesigning for humanlike interaction
29mAvoiding the Chatbot Trap Designing AI Interfaces That Unlock Real Business Value