Generative AI and LLMs have been adopted rapidly for mental health applications and in ways that impact mental health. These technologies are viewed as transformative and promising, yet a number of high-profile tragedies have drawn attention to ethical and safety concerns with the development and deployment of generative AI on mental health, such as episodes of psychosis, domestic violence and suicide. This session addresses: issues of appropriate use of Gen AI for mental health and concerns regarding use among different groups; key ethical issues, such as safety, transparency, bias and fairness; as well as policy recommendations.
- Ehsan Adeli, PhD
- Luca Belli, PhD
- Jodi Halpern, MD, PhD
- Nicole Martinez-Martin, JD, PhD
Coverage
Attend
Neuroethics 2026
Building Bridges and Cultivating Community
Thursday, April 16, 2026
This session was part of the Neuroethics 2026 conference organized by the INS and hosted by the Stanford Center for Biomedical Ethics at the Bechtel Conference Center at Encina Hall.
Agenda
- Presentations / 20 minutes
- Panel discussion / 15 minutes
- Audience Q&A / 25 minutes
Description
Generative AI and LLMs have been adopted rapidly across a range of societal institutions, including for mental health applications and in ways that impact mental health. Generative AI is viewed as transformative for psychiatry and mental health, with the capability of improving diagnosis, treatment and access. In particular, Gen AI is presented as a promising avenue for dealing with the existing shortages of mental health resources and care professionals — aiding with administrative burdens by writing patient notes, providing decision-making support and patient educational materials, and even being used to provide therapy. At the same time, a number of high-profile tragedies have drawn attention to ethical and safety concerns with the development and deployment of generative AI on mental health, such as episodes of psychosis, domestic violence and suicide.
The panelists addressed issues of appropriate use of Gen AI for mental health and concerns regarding use among different groups; key ethical issues, such as safety, transparency, bias and fairness; as well as policy recommendations.
Speakers
Ehsan Adeli, PhD
Stanford University
Ehsan Adeli is an Assistant Professor of Psychiatry & Behavioral Sciences and, by courtesy, of Computer Science and Biomedical Data Science at Stanford University. He also directs the Stanford Translational AI (STAI) in Medicine and Mental Health Lab focuses on developing Translational Artificial Intelligence (AI) algorithms in medicine and mental health, leveraging recent advancements in AI, computer vision, ambient intelligence, and computational neuroscience.. His research lies at the intersection of machine learning, computer vision, medical image analysis, and computational neuroscience.
Luca Belli, PhD
AI safety Lead, Spring Health
Luca Belli is the AI safety Lead at Spring Health. He works at the intersection of tech and policy, focusing on how algorithms shape culture and behavior, and examining the role and limits of AI systems such as chatbots within mental health journeys. He co-founded Twitter’s Machine Learning Ethics, Transparency, and Accountability team and is now lead AI Safety at Spring Health while serving as a Non-Resident Research Fellow at UC Berkeley. His recent work includes NIST Red Teaming, international standards, and advising the European Commission on algorithmic transparency.
Jodi Halpern, MD, PhD
Co-Director, Kavli Center for Ethics, Science and the Public
Professor Halpern is a leading thinker and speaker on the ethics of innovative technologies including gene editing, AI, and virtual/augmented reality (VR/AR). While many raise ethical concerns, Halpern brings decades of ground-breaking research in psychiatry, philosophy, and decision-making in science and medicine to provide road maps to ethical practice. She guides organizations to develop cultures of innovation that respect and protect human rights.
Nicole Martinez-Martin, JD, PhD
Stanford University
Nicole Martinez-Martin is an Assistant Professor of Pediatrics in the Center of Biomedical Ethics at Stanford University. Her background is in law, social sciences, and comparative human development. Her research interests include the examination of the impact of new technologies on marginalized populations, with a focus on equity and privacy. Her recent work in bioethics and neuroethics has centered on the ethics of AI and digital health technology, with grant-funded projects on digital phenotyping, digital mental health tools, and ambient intelligence.
Resources
Julia E.H. Brown, Jodi Halpern. AI chatbots cannot replace human interactions in the pursuit of more inclusive mental healthcare. SSM – Mental Health, Volume 1, 2021, 100017. (PDF) https://doi.org/10.1016/j.ssmmh.2021.100017
Keywords: generative AI, mental health, cultural anthropology, policy
Recording
A recording of this session will be made available to registered attendees.