New York Governor Kathy Hochul on Wednesday officially broke ground on the permanent home of Empire AI at the University at Buffalo, launching the next phase of Empire AI Gamma — the full-scale supercomputing facility the state says will cement its academic computing network as the most powerful AI research platform in American higher education.
The facility, expected to be completed by early 2028, will dramatically expand the computing power available to researchers at New York's leading public and private universities. Empire AI has grown into a more than $500 million public-private initiative since Hochul first announced it in her 2024 State of the State address, and it already supports research in medicine, climate science, advanced manufacturing and other fields.
A First-in-the-Nation Consortium
Launched in April 2024, Empire AI is a first-in-the-nation consortium of leading public and private research institutions that gives academic researchers access to world-class AI computing power no single university could afford independently. Just six months later, the state launched Empire AI Alpha, made possible by a philanthropic contribution from the Simons Foundation, giving hundreds of researchers immediate access to advanced computing resources.
The initiative's second phase arrived in August 2026, when Empire AI Beta — a $40 million supercomputer — became fully operational at the University at Buffalo, delivering 11 times the AI training capacity, 40 times the inference performance and eight times the data storage of Alpha. Empire AI Board Chair Tom Secunda said the consortium now counts more than 1,500 registered users with hundreds of research projects already underway.
"Two years ago, we made a commitment to put New York at the forefront of the AI revolution — while ensuring safe and responsible AI development is a top priority — and today we are delivering on that promise," Hochul said at the groundbreaking. "Empire AI is already giving our brightest researchers the most powerful academic AI computing resources in the country, and now we're building the permanent home that will take this initiative to an entirely new level."
Secunda framed the project as a national model: "Empire AI was an ambitious idea, and today it is a national model for putting world-class computing power in the hands of academic researchers." Marilyn Simons, co-founder of the Simons Foundation, said many of the greatest scientific breakthroughs begin with a question no one has been able to answer, and that Empire AI gives more scientists the tools to tackle those questions.
Paired With AI Safety Research
The groundbreaking comes just days after Hochul launched the New York Center for AI Responsibility and Research at Binghamton University, a first-of-its-kind initiative bringing together more than 80 scientists to independently study the safety, security and performance of advanced AI systems using Empire AI's computing infrastructure. State officials also pointed to New York's RAISE Act on frontier AI accountability as part of a broader strategy that pairs computing investment with oversight.
That combination — frontier-class academic compute alongside an institutional home for independent AI safety research — has become a template other states are watching. While federal AI policy remains contentious, states have increasingly taken the lead on both funding and regulating the technology, and New York has positioned itself at the front of that wave.
Why Academic AI Compute Matters
The push reflects a growing concern in the research community: as frontier AI development concentrates in a handful of companies, university researchers risk being priced out of the field entirely. Training and evaluating modern AI models requires compute budgets measured in the tens or hundreds of millions of dollars — resources that, until initiatives like Empire AI, existed almost exclusively inside private labs.
By pooling state funds, philanthropy and institutional contributions, Empire AI lets scientists at SUNY schools and private universities tackle bigger questions, accelerate research timelines and pursue discoveries that would otherwise be beyond the reach of individual institutions, state officials said. SUNY Chancellor John B. King Jr. said the consortium is driving the state's leadership in research that advances the public good, from accelerating the search for cures and treatments for disease to other applications.
The state says the new facility will also set new standards for energy efficiency and environmental responsibility, though detailed specifications for Gamma's hardware configuration have not yet been published.
With the permanent facility now under way, the question for New York is whether public academic compute can keep pace with the exponential growth of private AI infrastructure. For the 1,500 researchers already using the system, Gamma's early-2028 completion cannot come soon enough. Follow the latest AI developments in research infrastructure and public AI investment at AI Buzz Wire.
Stay Ahead of AI
The race for AI compute is reshaping research and industry alike. Get every update, plus the latest AI developments from every major lab, at AI Buzz Wire.
Read more AI news →