Emerald AI closed a $150 million Series A round at a $1.05 billion valuation to fund the development of power-flexible data centers.

Building new transmission infrastructure in the U.S. takes a decade or more, and the International Energy Agency projects that data centers will account for nearly half of U.S. electricity demand growth through 2030. Because the grid cannot expand quickly enough to meet AI’s growing power demands, Emerald AI is pursuing an alternative solution: transforming data centers into flexible grid assets rather than relying on additional transmission lines. The company has announced a $150 million Series A financing round at a $1.05 billion valuation. Oversubscribed and co-led by Energize Capital and DCVC, the round brings Emerald AI’s total funding to over $220 million.

Emerald AI’s core platform, Emerald Conductor, dynamically orchestrates a facility’s AI computing workloads alongside onsite energy resources, including batteries and generation systems. The software adjusts power consumption during periods of grid stress while safeguarding the performance of critical workloads. By shifting data centers from grid liabilities to reliable grid participants, Emerald AI estimates that widespread adoption could unlock more than 100 gigawatts of untapped capacity on the existing U.S. grid without requiring new infrastructure.

Over the past year, Emerald AI completed five commercial demonstrations across Arizona, Illinois, Virginia, Oregon, and London, collaborating with partners including NVIDIA, EPRI, Oracle, Nebius, and National Grid. The company has now transitioned from pilot testing to commercial scaling, deploying its technology at a fully operational data center in California that successfully demonstrated grid-responsive flexibility during peak demand periods. Additionally, Emerald AI partnered with Silicon Valley Power on the nation’s first Flexible Load Interconnection Program, which grants data centers expanded grid access in exchange for verified load flexibility. In Manassas, Virginia, the company is working with Digital Realty and NVIDIA to commission the world’s first power-flexible AI factory: the nearly 100-megawatt Vera Rubin AI Research Factory. Developed in collaboration with EPRI, Dominion, and PJM Interconnection, the facility is expected to come online later this year.

Founder and CEO Dr. Varun Sivaram is a Rhodes and Truman Scholar who holds a PhD in condensed matter physics from Oxford University. Prior to founding Emerald AI, he served as Managing Director for Clean Energy at the U.S. State Department, where he established the First Movers Coalition, and held chief strategy and technology leadership positions at Danish energy company Ørsted and Indian renewable developer ReNew Power. He is also the author of *Taming the Sun*, a book examining the future of solar energy. “We founded Emerald AI on the conviction that the intelligence driving the AI revolution could solve its own greatest bottleneck: power,” Sivaram said.

Founded in November 2024, Emerald AI initially secured a $24.5 million seed round led by Radical Ventures in July 2025, followed by a seed extension led by Lowercarbon Capital. The current Series A attracted participation from twelve Fortune Global 500 companies, including NVIDIA, Samsung Ventures, Siemens, Aramco Ventures, Salesforce Ventures, GE Vernova, and RWE. These investors also serve on the company’s Strategic Advisory Board, collaborating directly with Emerald AI on product integration and deployment. “The binding constraint on AI is no longer chips or capital; it is power, and software is the fastest way through it,” said John Tough, Managing Partner at Energize Capital. Zachary Bogue, Co-Founder and Managing Partner at DCVC, noted that the platform “makes flexibility a permanent feature of how data centers are powered.” The newly raised capital will be directed toward scaling commercial deployments globally.

Competitors are addressing the power bottleneck through divergent strategies. Stanford spinoff GridCARE recently raised $64 million to develop software that manages grid-to-data-center interconnection, mirroring Emerald AI’s approach. Meanwhile, XLR8 and VoltaGrid circumvent grid constraints entirely by deploying onsite generation, securing $1.75 billion and $1 billion, respectively. Google has pursued a third path, acquiring Intersect Power for $4.75 billion last December to vertically integrate and directly own power generation assets.