Research · Track 03
UX and UI in Agentic Applications
The interaction layer is the adoption bottleneck, and nobody owns it.
6–10 contributors
Open — seeking a lead
5–15 hrs/week contributing · 20+ hrs/week core
Why now
Context
AI is moving from chat to ambient, agentic and multi-device. The interaction layer — how a person delegates, oversees, interrupts and recovers — is where adoption and trust are actually won or lost. It is also the layer with the least published, reusable research.
This is the track most likely to produce something a working team uses next week. That is a feature, not a compromise — an open pattern library with reference implementations is a public good, and it is cheap to ship relative to its usefulness.
Sub-themes
Scope
- Interaction patterns — delegation, oversight, interruption, recovery
- Trust calibration and transparency in autonomous interfaces
- Multimodal UX across voice, gesture, gaze and ambient computing
- Designing for Indic users — language, literacy and accessibility
- Evaluation frameworks for agentic UX covering task, trust and cognitive load
- Open design patterns with reference implementations
Year one deliverables
Output
- One to two papers on patterns and evaluation
- An open design pattern library
- User studies across two to three device classes
If the pattern library is not being cited or forked outside IIDA within two quarters of release, it becomes a maintenance-only artefact rather than an active track.