Proceedings of the ACM on Human-Computer Interaction, Journal Year: 2025, Volume and Issue: 9(2), P. 1 - 36
Published: May 2, 2025
The mental health crisis in the United States spotlights need for more scalable training workers. While present-day AI systems have sparked hope addressing this problem, we must not be too quick to incorporate or solely focus on technological advancements. We ask empirical questions about how ethically collaborate with and integrate autonomous into clinical workplace. For these Human-Autonomy Teams (HATs), poised make leap domain, special consideration around construct of trust is order. A reflexive look toward multidisciplinary nature such HAT projects illuminates a deeper dive varied stakeholder considerations ethics trust. In paper, investigate impact domain---and ranges expertise within domains---on ethics- trust-related HATs health. outline our engagement 23 participants two speculative activities: design fiction factorial survey vignettes. Grounded by video storyboard prototype, AI- Psychotherapy-domain experts novices alike imagined TEAMMAIT, prospective system psychotherapy training. From inductive analysis emerged 10 themes surrounding ethics, trust, collaboration. Three can seen as substantial barriers collaboration, where they would work an teammate that didn't meet ethical standards. Another five interrelated, context-dependent, variable factors collaboration teammate. final represent explicit role practices. conclude evaluating findings through lens Mayer et al.'s Integrative Model Organizational Trust discuss risks adapt models ability-, benevolence-, integrity-based These updates motivate implications integration work.
Language: Английский