Minion: A Technology Probe for Resolving Value Conflicts through Expert-Driven and User-Driven Strategies in AI Companion Applications
This addresses value conflicts in AI companion applications to improve user experience, but it is incremental as it builds on prior user complaints and focuses on specific interventions.
The paper tackled the problem of value conflicts between users and AI companions by developing Minion, a technology probe that combines expert-driven and user-driven strategies, achieving a 94.16% success rate in resolving conflicts across 274 tasks.
AI companions based on large language models can role-play and converse very naturally. When value conflicts arise between the AI companion and the user, it may offend or upset the user. Yet, little research has examined such conflicts. We first conducted a formative study that analyzed 151 user complaints about conflicts with AI companions, providing design implications for our study. Based on these, we created Minion, a technology probe to help users resolve human-AI value conflicts. Minion applies a user-empowerment intervention method that provides suggestions by combining expert-driven and user-driven conflict resolution strategies. We conducted a technology probe study, creating 40 value conflict scenarios on Character.AI and Talkie. 22 participants completed 274 tasks and successfully resolved conflicts 94.16% of the time. We summarize user responses, preferences, and needs in resolving value conflicts, and propose design implications to reduce conflicts and empower users to resolve them more effectively.