Key points are not available for this paper at this time.
Despite a growing body of research about the design and use of conversational agents, existing work has almost exclusively focused on interactions between an agent and a human. Less is known about how an agent is perceived and used during human-human conversation. We compared conversations between dyads using AI-assisted and standard messaging apps and elicited qualitative feedback from users of the AI-assisted messaging app through interviews. We find discrepancies between the AI assistant's suggestions and the conversational content, which is also reflected in participant interviews. Our results are used to suggest some areas for improvement and future work in AI-assisted communication.
Hohenstein et al. (Fri,) studied this question.