Redefining Team Processes in Human-AI Collaboration: A Mixed-Methods Study Across Team Phases

Abstract

Artificial Intelligence (AI) agents introduce new challenges and opportunities across team processes. This study examines the impact of AI on transition, action, and interpersonal phases, and explores team processes central to human-AI collaboration. We surveyed 632 global virtual team members and interviewed nine AI experts. Survey results reveal that AI is seen as most useful during transition phases, less in action, and least in interpersonal phases. Usefulness ratings declined over time across all phases, especially interpersonal, indicating unmet expectations. Experts valued AI in action phases but expressed concerns about losing control during transition phases. Extending an established team processes model, we identify six processes and practices that foster effective human-AI collaboration: AI introduction, expectation management, positive storytelling, change management, navigating role shifts, and social interaction. Our findings highlight the need for targeted strategies to support these processes and manage perceptions of AI in teams.

Ähnliche Arbeiten