Are We On Track? AI-Assisted Active and Passive Goal Reflection During Meetings

TL;DR

The paper tackles the problem of absent or evolving intentionality in meetings by introducing AI-assisted reflection as a real-time support mechanism. It employs a design probe approach with two prototypes—Ambient Visualization and Interactive Questioning—fed by real meeting data from 15 knowledge workers to study how reflection can be facilitated during meetings. Key findings show goal clarification as foundational, with passive ambient feedback improving focus and active questioning driving immediate reflection and action but risking disruption; three design dimensions—what to reflect on, when to reflect, and who should reflect—framing practical design and implementation. The work offers design guidelines for content tailoring, timing-sensitive intervention strength, democratic input versus efficiency, and user control, with implications for future AI-enabled meeting tools that promote deliberate, goal-aligned collaboration throughout the meeting lifecycle.

Abstract

Meetings often suffer from a lack of intentionality, such as unclear goals and straying off-topic. Identifying goals and maintaining their clarity throughout a meeting is challenging, as discussions and uncertainties evolve. Yet meeting technologies predominantly fail to support meeting intentionality. AI-assisted reflection is a promising approach. To explore this, we conducted a technology probe study with 15 knowledge workers, integrating their real meeting data into two AI-assisted reflection probes: a passive and active design. Participants identified goal clarification as a foundational aspect of reflection. Goal clarity enabled people to assess when their meetings were off-track and reprioritize accordingly. Passive AI intervention helped participants maintain focus through non-intrusive feedback, while active AI intervention, though effective at triggering immediate reflection and action, risked disrupting the conversation flow. We identify three key design dimensions for AI-assisted reflection systems, and provide insights into design trade-offs, emphasizing the need to adapt intervention intensity and timing, balance democratic input with efficiency, and offer user control to foster intentional, goal-oriented behavior during meetings and beyond.

Figures (6)

• Figure 1: Ambient Visualization: The visualization evolves over time as users observe it at different moments during the meeting (e.g., panels A-C), with the content becoming richer as the discussion progresses
• Figure 2: Interactive Questioning: (1) AI pops up a question with three options at the time when AI identifies a reflective discussion is needed. (2) If the majority votes for reflection, then it proceeds. (3) AI listens to and summarizes the goal-oriented reflective discussion.
• Figure 3: Study overview. The diagram illustrates the overall study workflow, including data-preprocessing (A), system setups, and the workflow of simulating real-time AI-assisted reflection during the session (B), probe session procedure (C), and data analysis (D).
• Figure 4: Design dimensions---what to reflect on, when to reflect, and who should reflect---and implementation considerations for accommodating users' needs for in-meeting reflection.
• Figure 5: (A) 'Strong' intermittent interventions are much more direct and can provoke immediate action. However, they risk becoming disruptive if not timed properly. (B) 'Light' ambient interventions are more subtle. However, when subjective assistance needs are high, they may fail to capture attention. Conversely, when assistance needs are very low, even a light ambient intervention may unnecessarily add to users' cognitive load.