Skip to main content
Preprint Credible — Major labs, established NGOs, reputable named-author preprints

Open Problems in Chatbots and Mental Health

Position paper reviewing what its authors consider the most critical open problems for researchers and policymakers on general-purpose chatbots and mental health. It groups them into controlled evaluation of chatbots, linking studies to real-world evidence, and evidence-based policy.

Publisher

Stanford University, University of Chicago, Carnegie Mellon University, Vals AI, Harvard University, University of Minnesota, Vanderbilt University and University of Texas at Austin authors (arXiv preprint)

Published

7 Nov 2026

Added

4 days ago

DOI

—

Key Findings

  • Argues that conventional benchmarks relying on static datasets cannot establish chatbot safety, and that novel simulation methods for multi-turn evaluation need their own validation.
  • Names construct and ecological validity, multi-turn evaluation, and connecting evaluations to improvements in chatbot behavior as the tractable evaluation problems.
  • Calls for study of deployed products, including user interfaces and human review and escalation practices for welfare concerns, while noting that sharing real chat logs raises significant privacy concerns.
  • Proposes policy levers: user protections, standards of care, protected researcher access to chat logs, data sharing between institutions, incident reporting and post-deployment monitoring.

Methodology Notes

Preprint submitted to arXiv (cs.CL) on 2026-11-07 (arXiv:2611.10997v1); not peer-reviewed. Position and review paper with no new empirical data. Metadata and abstract read from the arXiv abstract page; author affiliations read from the first page of the PDF.

Authors

Jared Moore, Jacy Reese Anthis, William Agnew, Andrea Mock, Kevin Klyman, Sanmi Koyejo, Nick Haber, Stevie Chancellor, Eric Lin, Ryan Louie, Desmond C. Ong

Tags

position-paperresearch-agendaevaluation-validityincident-reporting

Cite This

APA

Jared Moore et al. (2026). Open Problems in Chatbots and Mental Health. Stanford University, University of Chicago, Carnegie Mellon University, Vals AI, Harvard University, University of Minnesota, Vanderbilt University and University of Texas at Austin authors (arXiv preprint). https://arxiv.org/abs/2611.10997