Here's more on our initial strategy behind these grants, and other grants we're investigating in Public Data for Health, that may be of interest to researchers: Our starting hypotheses for areas OpenAI Foundation support can be most useful are: connected data, scarce data, and direct data. Datasets we fund will usually have one or two of these properties, described in more detail in the next tweets, and in rare cases will share all three properties. Across all datasets, we believe making data broadly accessible, while ensuring privacy whenever that relates to patient data, will become increasingly important. This will allow researchers to build on each other’s work, and scientific progress to emerge from unexpected places.

3. Direct data: Measuring biological and clinical states closest to what matters 🔬 Biology has spent much of its history measuring what was easy or cheap to measure. For instance: measuring individual cell data over human clinical data, or RNA expression rather than proteins. These proxies have been enormously useful, but they can also become detached from the goals of understanding biology, and of preventing and treating disease. In cases like this, we want to give researchers the resources they need to assemble datasets that measure what matters most. It may also be the case that the necessary instruments to measure “biological truth” do not yet exist, or are prohibitively expensive to use at scale. Within this program, we hope to support both initiatives – new data, and new tools. For example, our grant to @UNC to establish the Initiative for Generative Immunotherapy will allow the team at UNC Lineberger and UNC Health to produce the human-grounded data necessary to improve the next generation of cancer vaccines. That means measuring tumor cells’ surface proteins directly, as well as patients’ T cell responses, rather than “only” sequencing the tumor. This data is expensive to create; by doing it comprehensively this time, and making the results open, we hope all future cancer vaccine efforts will be able to benefit. In particular, if this project is successful, future vaccine candidates will be able to enter human dosing with a stronger set of targets. That should improve their chances of working, and increase what can be learned when they do not.