Luster presented the virtual poster, “In-Session AI Reflection and Empathy: Pilot Findings and Practice Implications,” on August 8 during a virtual poster session organized by the Society for the Advancement of Psychotherapy, part of APA Division 29. The study investigated whether researcher-designed AI personas could create a consistent, transcript-rich environment in which participants practiced skills associated with empathy. These included viewing a situation from another person’s perspective, identifying emotions, responding with validation and finding language that could support relational repair.

Ten adults took part in two facilitated encounters with the AI personas. Rather than interacting with a general-purpose chatbot without supervision, participants engaged with AI systems constructed to present complex relational experiences. The facilitated format was central to the research design: participants were encouraged to slow down, consider competing viewpoints and examine how their responses affected the direction of the conversation. Researchers then reviewed the transcripts to identify patterns in communication and emotional development across the interactions.

The study used an eight-item empathy self-assessment before and after the encounters. Average scores rose from 23.10 before the sessions to 25.80 afterward, representing a mean increase of 2.70 points. Nine of the 10 participants recorded higher post-experience scores. The researchers reported a within-person effect size of dz = 0.89, a statistic that indicates a relatively large difference in this small sample. Yet the result should be interpreted cautiously because the pilot had no control group and cannot establish that AI exposure itself caused a lasting increase in empathy.

The researchers also examined how participants’ communication unfolded over time rather than focusing only on their final answers. Their analysis tracked perspective-taking, empathic concern, emotional labeling, curiosity, continuity across the conversation and meaning-making. Nine participants reached the third stage of the researchers’ relational progression framework, defined as disclosure of an underlying emotional wound. According to the reported qualitative findings, participants showed greater emotional clarity, more openness to alternative interpretations and less certainty that another person’s actions were intentionally hostile.

One potentially important observation was that progress appeared to depend on accumulated relational safety rather than a single ideal response. In other words, participants did not necessarily demonstrate empathy by producing one perfectly worded statement. Instead, their conversations appeared to develop through repeated moments of curiosity, emotional recognition and validation. This makes the transcripts particularly valuable for educators and supervisors, who could use them to examine how a person’s communication changes throughout an interaction and where an exchange begins to become more constructive or defensive.

The prospect of using AI for this purpose comes as more people turn to conversational systems while dealing with conflict, loneliness and emotional distress. Researchers and clinicians, however, have limited evidence about when AI-assisted reflection is useful, when it may be misleading and how it should be integrated into professional practice. Luster’s proposed model positions generative AI as an adjunctive process tool: a structured environment for rehearsal and reflection that remains subject to human interpretation and oversight.

That distinction is critical because AI-generated responses can be inaccurate, overly confident or insensitive to context. The presentation emphasized informed consent, privacy and confidentiality, clear professional boundaries, documentation of AI-assisted exercises and preservation of participant autonomy. Clinicians, instructors and supervisors would remain responsible for deciding whether AI-generated material is relevant, safe and appropriate for a particular person or learning situation. They would also need to review the output critically rather than treating it as an objective assessment of emotion or interpersonal behavior.

The researchers identified situations in which AI-assisted exercises may be inappropriate, including high-acuity clinical conditions, paranoia, impaired reality testing, heightened suggestibility and circumstances involving substantial privacy vulnerability. These concerns are especially significant because a simulated persona can appear responsive and emotionally fluent without possessing human understanding or clinical judgment. Future studies will need larger and more diverse samples, control conditions, longer follow-up periods and independent testing of the relational progression framework. For now, the pilot offers an intriguing but preliminary signal that supervised AI encounters might help people practice difficult conversations before applying those skills in real relationships, workplaces or clinical training. Its strongest message is not that machines can create empathy on their own, but that carefully designed technology may give human professionals another instrument for teaching, observing and refining it.

Subject of Research: People; psychological science; interpersonal skills and empathy practice

Article Title: Can Supervised AI Personas Help People Practice Empathy? Pilot Study Offers a Cautious Signal

Web References: APA 2026 program abstract and poster

References: Luster, R. “In-Session AI Reflection and Empathy: Pilot Findings and Practice Implications.” Society for the Advancement of Psychotherapy, APA Division 29 virtual poster session, 2026.

Keywords: generative artificial intelligence, empathy, perspective-taking, counselor education, psychotherapy, interpersonal communication, emotional labeling, relational repair, AI ethics, psychological science

AI mental health applications, ranging from mood-tracking tools to chatbots that simulate human therapists, have become increasingly popular. These technologies promise a revolution in accessibility by providing affordable, readily available support. However, as we embrace these innovations, it is essential to pause and consider the implications of relying heavily on AI when it comes to mental health care, particularly for vulnerable populations such as children. The intersection of AI technology and pediatric mental health care raises many ethical questions that necessitate careful exploration.

Current AI mental health applications primarily target adult users and operate largely in an unregulated landscape. This reality does not always resonate with the unique developmental needs of children, who require more than just symptom management; they need a supportive environment that takes into account their intricate social and familial context. As human beings, children navigate their emotional lives intertwined with their relationships, which AI lacks the capability to fully comprehend or replicate. Dr. Bryanna Moore, an assistant professor of Health Humanities and Bioethics, emphasizes the need for inclusive conversations regarding these technological solutions. According to Moore, our focus should include the distinct cognitive and social development stages children experience as they grow.

Another critical concern is the potential developmental impact on children’s social skills when they interact with mental health chatbots. Studies indicate that children may attribute human-like qualities to these machines, leading them to form attachments that might impede their ability to build genuine relationships with their peers and caregivers. This form of reliance on AI could stifle vital interpersonal skills and have long-term repercussions on their social development. In pediatric therapy, professionals always consider the context in which a child exists, recognizing that a family’s dynamics play a crucial role in their overall mental health.

The providers of mental health services aim to ensure the safety and well-being of children by integrating their family into the therapy process. Therapists actively observe and engage with a child’s social relationships, assessing risks and employing interventions when necessary. AI chatbots, however, lack the contextual awareness about a child’s environment and relationships that is vital for effective mental health interventions. Consequently, there is a missing opportunity for critical responses to situations where a child might be in danger or experiencing distress.

Moreover, concerns surrounding health equity arise when discussing the implementation of AI mental health tools. Experts highlight that AI’s effectiveness is strongly tied to the quality of data it is trained on. Without ensuring that diverse and representative datasets are utilized, AI applications run the risk of reinforcing existing disparities in mental health care. Dr. Jonathan Herington, a coauthor of the commentary alongside Moore, points out that marginalized communities often face compounded barriers in accessing traditional mental health services. Consequently, the introduction of AI chatbots could inadvertently position them as the sole means of support for these individuals, further entrenching inequities.

This is particularly true for children from lower socioeconomic backgrounds who already exhibit a heightened risk for adverse childhood events, such as neglect or exposure to domestic violence. These traumatic experiences can result in the necessity for comprehensive mental health support, yet accessing such treatment remains a significant challenge for many families. As Herington posits, while AI chatbots can serve as valuable supplemental resources, they must never be seen as substitutes for genuine human-led therapies.

As it stands, the AI mental health chatbot landscape is largely unregulated—a concern that demands immediate attention. To date, the U.S. Food and Drug Administration has approved only one AI-driven mental health app designed for treating major depression in adults. The absence of thorough regulations leaves the door open for misuse and inequity in training data, compounded by gaps in user access. This absence underscores the urgent need to formulate standards that guarantee ethical usage in AI development while elevating the importance of human oversight.

Both experts agree that their aim is not to dismiss the potential of AI tools but rather to advocate for a thoughtful and balanced approach in their deployment—particularly when addressing the complex nuances of children’s mental health. Moore urges the importance of having a dialogue centered around the potential risks associated with AI, especially as developers continue exploring the intersection of technology and emotional support.

Furthermore, Moore and Herington, alongside Dr. Şerife Tekin, maintain that engagement with developers is vital. Understanding the specific methodologies applied in creating AI-based therapy chatbots can enrich the discourse on ethical considerations and safety factors inherent to such technology. By collaborating with researchers, pediatricians, parents, and the children themselves, developers can ensure that the design of AI tools aligns with sound evidence-based practices.

To adequately harness the benefits of AI in mental health care, an ethical framework should be established. This framework would not only guide AI developers in creating responsible tools but also facilitate an environment in which the emotional needs of children are prioritized. The collaborative efforts of healthcare professionals, ethicists, data scientists, and families can pave the way for AI that resonates with the vulnerabilities and developmental needs of children while addressing potential ethical dilemmas head-on.

Investing in thoughtful discussions and partnerships, particularly involving the pediatric population, can lead to a future where AI acts as an ally rather than a surrogate, fostering frameworks that embrace holistic mental health care for younger generations. As the conversation around AI in mental health care evolves, we must remain critical and conscientious about the pathways we choose to pursue, ensuring we support children’s growth while equipping them with the necessary tools to thrive emotionally and socially.

Subject of Research: People
Article Title: The Integration of Artificial Intelligence-Powered Psychotherapy Chatbots in Pediatric Care: Scaffold or Substitute?
News Publication Date: 10-Mar-2025
Web References: http://dx.doi.org/10.1016/j.jpeds.2025.114509
References: The Journal of Pediatrics
Image Credits: The Journal of Pediatrics

Keywords: Artificial intelligence; Mental health; Children; Medical ethics; Generative AI