OpenAI's $2M mental health research push draws over 1,000 applicants
Curated by the Inblix editorial team
OpenAI’s attempt to fund independent research into AI and mental health turned into a flood—more than 1,000 proposals poured in before the December deadline, making it one of the company’s largest ever calls for research grants. The program dangled up to $2 million to back projects that dig into the messy overlap between increasingly personal AI use and emotional well-being. That’s a lot of academics betting the chatbot in your pocket needs a better bedside manner.
The company isn’t starting from zero here. They’ve already published work like ‘Investigating Affective Use and Emotional Well-being on ChatGPT’ and the Healthbench benchmark. The goal with these grants was to push beyond internal tinkering. OpenAI specifically wanted interdisciplinary teams—technical researchers paired with mental health experts or people with lived experience—to produce concrete tools like datasets and evaluation rubrics, or nuanced insights into how mental health language varies across cultures and slips past current content classifiers.
The sheer volume of submissions, which OpenAI called ‘high-quality entries from both established and emerging researchers,’ signals a field scrambling to catch up with user behavior. People are already using these models for deeply personal conversations, whether the safety guardrails are ready or not. The company noted they’re actively exploring ways to expand the program, a nod to the fact that a one-time $2 million pot barely scratches the surface of the questions piling up.
What remains unspoken is the pressure this puts on the selected teams. Deliverables need to be actionable, directly informing OpenAI’s safety work. That’s a tight leash for independent research—funded by the company whose products are the subject of the investigation. It’s a dynamic worth watching: can you produce genuinely critical insights when your grant depends on being useful to the model builder? The answer will start to emerge as those funded proposals get to work.
💡 Key Takeaways
- OpenAI's $2 million fund attracted over 1,000 applicants, making it one of the company's largest-ever research grant calls.
- The program required interdisciplinary teams to produce specific deliverables like datasets or evaluations, not just theoretical papers.
- OpenAI is actively looking to expand the program after the unexpected volume of interest, signaling a recognition that initial funding was insufficient.
- The reliance on OpenAI for funding raises questions about the independence of research designed to directly inform the company's own safety work.
Keep reading: See related articles below for more coverage on this topic.
Get smarter about AI
The sharpest AI news, curated daily. Delivered free to your inbox.