Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

IRL Dittos: Embodied Multimodal AI Agent Interactions in Open Spaces

Seonghee Lee, Denae Ford, John Tang, Sasa Junuzovic, Asta Roseway, Ed Cutrell, Kori Inkpen

TL;DR

This work investigates embodied AI proxies that represent remote colleagues in open-office spaces to enable real-time social interactions. It introduces IRL Ditto, an embodied, multimodal agent that speaks in the Source's voice, uses proxemic cues, and references personal knowledge to engage passersby in hallway encounters. A four-day, IRB-approved field study with 21 participants reveals that relationship familiarity with the Source modulates effectiveness, while social repair and grounding are critical for trust and natural dialogue. The findings highlight the potential of embodied agents to enrich distributed teams' workplace dynamics and outline concrete avenues for improving grounding, responsiveness, and identity signaling in future iterations.

Abstract

We introduce the In Real Life (IRL) Ditto, an AI-driven embodied agent designed to represent remote colleagues in shared office spaces, creating opportunities for real-time exchanges even in their absence. IRL Ditto offers a unique hybrid experience by allowing in-person colleagues to encounter a digital version of their remote teammates, initiating greetings, updates, or small talk as they might in person. Our research question examines: How can the IRL Ditto influence interactions and relationships among colleagues in a shared office space? Through a four-day study, we assessed IRL Ditto's ability to strengthen social ties by simulating presence and enabling meaningful interactions across different levels of social familiarity. We find that enhancing social relationships depended deeply on the foundation of the relationship participants had with the source of the IRL Ditto. This study provides insights into the role of embodied agents in enriching workplace dynamics for distributed teams.

IRL Dittos: Embodied Multimodal AI Agent Interactions in Open Spaces

TL;DR

This work investigates embodied AI proxies that represent remote colleagues in open-office spaces to enable real-time social interactions. It introduces IRL Ditto, an embodied, multimodal agent that speaks in the Source's voice, uses proxemic cues, and references personal knowledge to engage passersby in hallway encounters. A four-day, IRB-approved field study with 21 participants reveals that relationship familiarity with the Source modulates effectiveness, while social repair and grounding are critical for trust and natural dialogue. The findings highlight the potential of embodied agents to enrich distributed teams' workplace dynamics and outline concrete avenues for improving grounding, responsiveness, and identity signaling in future iterations.

Abstract

We introduce the In Real Life (IRL) Ditto, an AI-driven embodied agent designed to represent remote colleagues in shared office spaces, creating opportunities for real-time exchanges even in their absence. IRL Ditto offers a unique hybrid experience by allowing in-person colleagues to encounter a digital version of their remote teammates, initiating greetings, updates, or small talk as they might in person. Our research question examines: How can the IRL Ditto influence interactions and relationships among colleagues in a shared office space? Through a four-day study, we assessed IRL Ditto's ability to strengthen social ties by simulating presence and enabling meaningful interactions across different levels of social familiarity. We find that enhancing social relationships depended deeply on the foundation of the relationship participants had with the source of the IRL Ditto. This study provides insights into the role of embodied agents in enriching workplace dynamics for distributed teams.
Paper Structure (30 sections, 2 figures, 1 table)

This paper contains 30 sections, 2 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. AI Proxies in Collaborative Workspaces
  4. Fostering Social Connection through Life-Sized AI and Digital Representation
  5. Proxemic Interactions and Digital Agents in Open Spaces
  6. System Overview
  7. Agent Representation
  8. Setup and Devices
  9. Agent Interaction Transition Sequence
  10. Detecting a Person in the Scene
  11. Entering the Engaged State
  12. Engaging in the Conversation
  13. Leaving, Pausing, and Ending the Conversation
  14. Summarizing the Conversation
  15. Study Design and Procedure
  16. ...and 15 more sections

Figures (2)

  • Figure 1: At the start, the IRL Ditto begins in the "Not Engaged" state, where a Kinect camera and UWB sensors detect individuals and log proximity and identification information within the journal. The model then makes an engagement check decision based on this logged information. If there is an interaction intent from the participant, the agent transitions to the "Engaged" state. During engagement, the conversation continues, and periodic disengagement checks are performed. If disengagement is detected, the system returns to the "Not Engaged" state.
  • Figure 2: Signage indicating the active IRL Ditto Zone.