Study maps UX design space for LLM-based computer use agents via two-phase research. Phase 1 reviewed systems and interviewed eight UX/AI practitioners to create taxonomy. Categories cover user prompts, explainability, user control, and more.
Key Points
- 1.Two-phase study on agent UX
- 2.Taxonomy from systems review and interviews
- 3.Focus on interaction design factors
Impact Analysis
Guides better UX for AI agents controlling computers, improving user trust and adoption. Informs future Apple AI tools for seamless human-agent interaction.
Technical Details
Agents interact via UI elements; taxonomy refined through practitioner insights. Explores user desires in agent collaborations.