The study, led by the University of Phoenix’s Center for Leadership and Entrepreneurial Research, tracked 10 adults as they engaged with AI-constructed personas designed to simulate complex relational scenarios. Participants underwent pre- and post-experience empathy assessments, with mean scores rising from 23.10 to 25.80. Nine of the 10 participants showed measurable improvement, often reaching higher levels of emotional disclosure during the sessions.
Despite these promising figures, researchers stressed that the technology is intended strictly as an adjunctive tool rather than a replacement for human clinicians or supervisors. The pilot results are preliminary and lacked a control group, meaning the findings demonstrate feasibility rather than proof of long-term behavioral change. Luster emphasized that the efficacy of such exercises relies heavily on professional oversight, particularly in managing privacy, informed consent, and the interpretation of AI output. Future studies are expected to expand to more diverse populations to determine if these early results can be replicated on a larger scale.

Comments (0)
No comments yet. Be the first!