The OpenAI Foundation — the charity arm of the artificial-intelligence behemoth — is on track to become one of the wealthiest non-profit organizations on the planet. And, with a 26% stake in OpenAI, which is seeking a valuation of around $1.4 trillion, it could even become the richest. The foundation was created in October 2025, after a tumultuous shake-up of OpenAI’s governance and structure. It has so far committed to giving away at least $25 billion.

This month, the foundation cut a round of cheques as part of a US$125-million commitment to fund the creation of freely available data for health and life-sciences research. The grants included $40 million for a project at the University of North Carolina at Chapel Hill to develop cancer vaccines; $500,000 for a pilot project to repurpose data from bankrupt biotechnology companies; and $15 million to an effort to better predict how drugs cross the blood–brain barrier and other membranes. Those awards follow a tranche of $100 million for Alzheimer’s disease research.

Jacob Trefethen, who previously worked at Coefficient Giving, a billionaire-funded foundation, is now head of life sciences at the OpenAI Foundation in San Francisco, California. He tells Nature that this is only the start, and the foundation plans to move fast.

Why is the OpenAI Foundation funding health and life sciences?

Our mission is to ensure AI benefits all of humanity. Part of the reason that we’re starting in the life sciences and health is that we think that’s one of the areas where AI can have the most benefit.

The non-profit Chan Zuckerberg Initiative announced in 2016 that its aim was to cure all disease by the end of the century. Does the OpenAI Foundation have a similar goal?

We intend to be ambitious. We have started with three areas in the life sciences to focus: AI for Alzheimer’s; public data for health; and high-burden diseases — which for us means any disease that leads to over half a million deaths around the world each year or similarly massive ill health.

We’re going to try and ask the question, what’s newly possible, for example, in tuberculosis — the largest infectious killer — that may help people who don’t currently have good-enough options. We have a kind of unifying philosophy and a hope to move quickly.

How do biological data fit in?

We think that the combination of intelligence and more observations of the world is what will deliver breakthroughs for people against disease. And those observations mean scientific data.

We think a lot of these problems that we’re working on, like Alzheimer’s, we collectively don’t know enough to deliver treatments that are excellent for patients. And so that requires more data of various forms to get there, probably. We want to support that underlying scientific infrastructure so that all researchers can build on it and all patients can benefit.

Is it a bit self-serving for the OpenAI Foundation to support projects that will generate data that could make OpenAI’s model more capable?