OpenAI has introduced a grant program to support independent research into the relationship between AI and mental health. The initiative commits up to $2 million to projects that investigate real-world risks, benefits, and applications of AI to enhance safety and well-being. The application window for these grants closed on December 19, 2025, with notifications for selected proposals issued by January 15, 2026.
Key Points
- OpenAI is awarding up to $2 million in grants for research on AI and mental health.
- The program supports independent safety and well-being research.
- Applications for the grants were open until December 19, 2025.
- Selected applicants were notified on or before January 15, 2026.
- Over 1,000 entries were received, making it one of OpenAI's largest calls for research grants to date.
- OpenAI is exploring ways to expand and build on this work due to high interest.
Context
According to OpenAI, the program aims to deepen understanding and accelerate innovation in the emerging field of AI and mental health. The organization seeks proposals that explore both the potential risks and benefits of AI in this domain, with a goal of building a safer and more helpful AI ecosystem. OpenAI emphasizes interdisciplinary research, encouraging collaboration between technical researchers, mental health experts, and individuals with lived experience. The grants are designed to support foundational work that strengthens OpenAI's own safety efforts and the broader field.
Successful projects are expected to produce clear deliverables such as datasets, evaluations, or rubrics, or generate actionable insights. These insights could include synthesized views from individuals with lived experience, descriptions of how mental health symptoms manifest in specific cultures, or research on language and slang used in mental health discussions that AI classifiers might miss. OpenAI's internal researchers and experts reviewed applications on a rolling basis.
Why It Matters
This grant program indicates a focus on understanding the societal implications of AI, particularly in sensitive areas like mental health. For builders and researchers, it highlights the importance of integrating ethical considerations and real-world impact assessments into AI development, moving beyond purely technical metrics to address human well-being and safety.
What To Do
- Note the types of deliverables and insights OpenAI sought from successful projects.
- Consider how interdisciplinary collaboration can inform AI safety work.
- Watch for future announcements regarding expansions of this research program.
Keep Exploring
/atlas/claude-family /atlas/gpt-family /techniques/system-user-separation /techniques/ptcf