Suchergebnisse

AbstractGiven that AI is becoming an increasingly active participant in work teams, this study explores how team trust emerges in human–AI teams compared to human–human teams. Adopting a multi‐level approach, we conducted two experimental studies (NStudy1 = 247 two‐member teams and NStudy2 = 106 three‐member teams, 828 individuals overall) and investigated how team composition (with AI or human team members) impacts interpersonal trust (affective and cognitive) and thus team trust. In two‐member teams, interpersonal trust via perceived trustworthiness and not via perceived similarity was lower in human–AI teams compared to human–human teams. Exploratory findings showed that team identification and cognitive interpersonal trust were also lower in two‐member human–AI teams than in human–human teams. However, in three‐member teams, we found no differences in team trust via interpersonal trust between the two team types. Instead, our findings revealed that perceived trustworthiness and perceived similarity increased interpersonal trust and, in turn, team trust for both team types. With this research, we showed that underlying theories and evidence of team trust in human‐only teams can enhance understanding of human–AI teams, though the results indicated certain differences that call for further investigation.

Investigating the team adaptation process in two laboratory experiments ( N = 144 teams, n = 504 participants), we found no benefits for teams with team adaptation experience (vs. without) nor for teams with external team adaptation experience (vs. with internal experience). Collective experience under routine and nonroutine conditions seems to provide teams with the resources to adapt. We further found that executing the team adaptation process did not always lead to high team performance; different team performance requirements might explain these findings. We discuss how our experimental findings can extend our understanding of team adaptation toward new boundary conditions.