The cover above was generated using DALL·E.

Paper icon

Download the proposal

Abstract. The new generation of generative AI models offers interactive opportunities that may fulfill long-standing aspirations in human-computer interaction and open doors to new forms of interaction that we have yet to imagine. The UIST community has a unique vantage point that can lead to critical contributions in envisioning a future of interactive computing that appropriately leverages the power of these new generative AI models. However, we are only just beginning to understand the research area that exists at the intersection of interaction and generative AI. By bringing together members of the UIST community interested in this intersection, we seek to initiate discussions on the potential of generative AI in architecting new forms of interactions. Key topics of interest include the exploration of novel categories of interactions made possible by generative AI, the development of methods for enabling more powerful and direct user control of generative AI, and the identification of model and architecture requirements for generative AI in interaction literature. The workshop will foster community building and produce concrete deliverables, including a research agenda, model/architecture requirements, and a simulated debate generated by a generative agent architecture.

Overview

Topics

  • Usage of generative AI: What are the novel categories of interactions that can be enabled by generative AI, and how do they contribute to the existing literature on interaction?
  • Controllability: How can we enable more powerful and direct user control of generative AI models and their outputs?
  • Model and architecture requirements: What assumptions or capabilities of generative AI models and their associated architectures do we need to consider in order to formulate new forms of interaction and establish reproducible evaluation techniques?

Agenda

  1. Session 1 -- Lightning talks: The first morning session of the workshop will be a 2-minute lightning talk from each participant.
  2. Session 2 -- Morning breakout session and spotlight talks: The second morning session will consist of a breakout session, where participants will form subgroups focusing on the topics mentioned in the introduction. This will be followed by a series of invited spotlight talks.
  3. Session 3 -- Afternoon debate: The afternoon session will feature a debate involving the generative agents acting as proxies of participants. While we hope this will be humorous and enjoyable in its own right, the hope is that hands-on experience trying to shape your own generative agent to represent you correctly will highlight the challenges with control of generative models.
  4. Session 4 -- Research agenda: In the penultimate session of the workshop, we will collaborate to create a research agenda that draws inspiration from the workshop.
  5. Session 5 -- Keynote: The final session will feature a keynote from Will Wright, renowned game designer and creator of The Sims and SimCity, and Lauren Elliott, co-designer of the Where In The World Is Carmen Sandiego series of games.

Joining the Workshop

Unlike typical SIGCHI workshops, workshops at UIST 2023 are open enrollment. There are limited spaces, offered first-come, first-served. Join the workshop by registering for it when you complete your UIST registration: https://uist.acm.org/2023/attending/#registration. Register early to ensure a space.

The workshop will take place in-person.


Contact

For any questions about the workshop, contact: interactivegenai@lists.stanford.edu.

Spotlight Talks

The slides from the three spotlight talks, given by Björn Hartmann (UC Berkeley), Jeffrey P. Bigham (CMU/Apple), and Maneesh Agrawala (Stanford), are available at the links below. Please cite these talks as described here.

Maneesh Agrawala. 2023. Designing Generative AI for Iteration. UIST Architecting Novel Interactions With Generative AI Models Workshop, Spotlight Talk (2023).

Jeffrey P. Bigham. 2023. How Might HCI Engage With Easy Access to Unintuitive Statistical Likelihoods of Things? UIST Architecting Novel Interactions With Generative AI Models Workshop, Spotlight Talk (2023).

Björn Hartmann. 2023. Expanding the Design Space of Generative AI-Powered Applications. UIST Architecting Novel Interactions With Generative AI Models Workshop, Spotlight Talk (2023).

Keynote

A profile image of Michael S. Bernstein

Will Wright
A profile image of Michael S. Bernstein

Lauren Elliott

Will is a renowned game designer and creator of The Sims and SimCity, and Lauren is the co-designer of the Where In The World Is Carmen Sandiego series of games. Leveraging their expertise in designing immersive open worlds, Wright and Elliott will explore the practical implications of incorporating generative AI into interactive experiences. They will discuss how these AI models can enhance user engagement, foster creativity, and facilitate collaborative storytelling. The speakers have confirmed their attendance.

Organizers

A profile image of Michael S. Bernstein

Michael S. Bernstein

Stanford University
A profile image of Joon Sung Park

Joon Sung Park

Stanford University

A profile image of Meredith Ringel Morris

Meredith Ringel Morris

Google DeepMind
A profile image of Saleema Amershi

Saleema Amershi

Microsoft Research
A profile image of Lydia Chilton

Lydia Chilton

Columbia University
A profile image of Mitchell Gordon

Mitchell L. Gordon

MIT and U of Washington




Citation.

@inproceedings{InteractionXGenAI,
    author = {Bernstein, Michael S. and Park, Joon Sung and Morris, Meredith Ringel and Amershi, Saleema and Chilton, Lydia and Gordon, Mitchell L.},
    title = {Architecting Novel Interactions With Generative AI Models},
    year = {2023},
    publisher = {Association for Computing Machinery},
    address = {New York, NY, USA},
    booktitle = {In the 35th Annual ACM Symposium on User Interface Software and Technology Workshop (UIST Workshop '23)},
    keywords = {generative AI, interaction design},
    location = {San Francisco, CA, USA},
    series = {UIST Workshop'23}
}
Architecting Novel Interactions With Generative AI Models. UIST 2023