UNC Lineberger Comprehensive Cancer Center announced Tuesday that it has received a $40 million grant from the OpenAI Foundation, money the Chapel Hill research center says will fund data collection aimed at making personalized cancer vaccines work better for patients. The gift lands the center at the center of a new nonprofit push to solve one of oncology's stubborn data problems: figuring out exactly which tumor targets a patient's immune system will actually attack.
A New Initiative Built to Measure, Not Just Predict
Under the grant, UNC Lineberger will establish the Initiative for Generative Immunotherapy to directly measure tumor surface proteins and patient T-cell responses across hundreds of tumors, bypassing the field's traditional reliance on tumor DNA sequencing alone, according to Unite.AI. As ABC7 Los Angeles reports, the collected data will be used to train improved AI-based methods for selecting tumor antigens for personalized vaccine therapies, and UNC Lineberger says it will make its findings broadly available to scientists worldwide.
The cancer-vaccine work includes Dr. Benjamin Vincent, an assistant professor in UNC's Division of Hematology-Oncology. Dr. Alex Rubinsteyn is an assistant professor of genetics and an expert in machine learning and computational medicine. Per ABC7's report, Rubinsteyn said personalized cancer vaccines are beginning to show clinical efficacy but many fail during development because they are too arbitrary. He said the goal is to use rich data and artificial intelligence to design more personalized and effective cancer vaccines.
Testing Multiple Vaccine Formulations in Breast Cancer Trials
Cancer vaccines are designed using tumor cells and immune cells drawn from a patient, and they work by training the immune response to recognize tumor surface proteins and attack cancer cells, the outlet's report notes. Tumor surface proteins are targets of particular interest for cancer-vaccine researchers because they offer a more direct read on what the immune system can actually see and respond to. UNC Lineberger researchers will compare multiple vaccine formulations in clinical trials evaluating their ability to raise tumor-specific immune responses in triple-negative breast cancer, an aggressive form of the disease that lacks estrogen, progesterone, and HER2 receptors, while metastatic breast cancer is expected to kill roughly 42,000 people annually in the United States, according to the Real Pink Podcast. Standard hormonal therapies do not work against triple-negative tumors, which is part of why researchers chose it as an initial testing ground.
Part of a Bigger Bet on Open Health Data
The UNC grant is one piece of a much larger rollout. The OpenAI Foundation launched its second science program, called Public Data for Health, on Tuesday, committing more than $125 million in initial grants to fund open-access scientific datasets, according to Unite.AI. The program aims to advance life science and disease research for the public good by supporting high-quality, open-access datasets for researchers and AI tools worldwide.
Jacob Trefethen, the Foundation's Head of Life Sciences, said AI-assisted cancer treatment research depends on having appropriate biological data, per ABC7's account. Unite.AI adds that the Foundation cited data showing roughly 90% of drug candidates fail in clinical trials, with difficulty predicting how drugs are absorbed and distributed.
Public Data for Health is the OpenAI Foundation's second initiative under its Life Sciences and Curing Diseases division, following the April 2026 launch of AI for Alzheimer's, for which the Foundation was working to finalize more than $100 million in grants across six research institutions, Unite.AI reports. The OpenAI Foundation is the nonprofit parent organization governing OpenAI Group PBC, and according to TNW, it controls the governing stake in that public benefit corporation following an October 2025 recapitalization that valued the nonprofit's equity at roughly $130 billion and anchored a $25 billion long-term pledge to health research and AI resilience. Guided by Board Chair Bret Taylor and Trefethen, the Foundation committed in the next year to invest at least $1 billion across life sciences and curing diseases, jobs and economic impact, AI resilience, and community programs, the same outlet reported.
Years of Groundwork at UNC
The push for better biological data builds on years of UNC research into why cancer vaccines have struggled. In a 2019 review published in Nature Reviews Cancer, UNC Lineberger researchers, with Dr. Christof Smith as first author, described target discovery and alignment as key challenges, according to the UNC Lineberger Comprehensive Cancer Center. Rubinsteyn has separately been involved in work involving computational tools and a Jaime Leandro Foundation grant, according to UNC's School of Medicine Department of Genetics.
UNC Lineberger says it will generate biological data at a scale researchers currently lack, with the new initiative positioned to feed that data back into AI models meant to guide future vaccine design. The center's clinical trial work in triple-negative breast cancer will test how directly measured protein and T-cell data can inform future vaccine design.
