Cited in the talk

  1. Leviathan, Y., Valevski, D., Kalman, M., Lumen, D., et al. (2026). Generative UI: LLMs are Effective UI Generators. Google Research, arXiv preprint 2604.09577. Paper · Project page

    The quote on The economics changed (from its conclusion), and generating interfaces as an emergent capability of newer models (section 3.1, tables 3 and 4) in How would this be implemented, and does it work today?

  2. Leviathan, Y., Valevski, D., Natchu, V., & Matias, Y. (2025, November 18). Generative UI: A rich, custom, visual interactive user experience for any prompt. Google Research blog. Post

    Google Search’s AI Mode building a custom interface for each query, on AI already generates UIs

  3. Karger, D. R., Bakshi, K., Huynh, D., Quan, D., & Sinha, V. (2005). Haystack: A Customizable General-Purpose Information Management Tool for End Users of Semistructured Data. Conference on Innovative Data Systems Research (CIDR 2005). Paper (PDF)

    David Karger’s quote on Apps can’t predict you

  4. Watson, L. (2024, November 8). AI and Accessibility: the Good, the Bad, and the Bollocks. Talk at FFConf 2024, Brighton. Video, from 11:15 · Talk page

    The clip we play on AI & accessibility today

  5. Appleton, M. (2024, May). Home-Cooked Software and Barefoot Developers. Talk at the Local-first Conference, Berlin. Talk page

    The barefoot developer drawing on By the people, for the people

  6. Apple. Accessibility: Vision (VoiceOver and Live Recognition). Page

    Narrated descriptions of the camera’s view, to try with VoiceOver, on AI & accessibility today

Tools our interviewees use

  1. Be My Eyes: sighted volunteers and AI describe what the phone’s camera sees. Site

    Molly

  2. Dragon NaturallySpeaking: speech recognition and voice control for Windows, by Nuance. Wikipedia

    Myron

  3. Ravelry: a community site for knitters and crocheters, whose redesign put Rose in bed for three days. Site

    Rose · Doesn’t a changing UI break screen readers?