GridCARE addresses one of the defining constraints of the AI era: electricity. As AI data centers strain power infrastructure, new capacity can take years to interconnect. GridCARE's premise is that significant capacity already exists on today's grid but sits underutilized, and that AI can find and activate it far faster than building new lines.
The company's Energize platform applies physics-based AI to model grid behavior and pinpoint near-term capacity, enabling 'AI factories,' utilities and energy providers to site and power large loads without waiting through multi-year queues. GridCARE frames this as a new category it calls 'power acceleration for AI.'
Founded out of Stanford's Doerr School of Sustainability, GridCARE's team blends energy and AI heavyweights: CEO Amit Narayan, co-founder Arun Majumdar (former Google VP of Energy and founding ARPA-E director), Liang Min from Stanford's Bits & Watts lab, and CTO Ram Rajagopal. That pedigree underpins its physics-informed approach to grid modeling.
In May 2026 GridCARE closed an oversubscribed $64 million Series A led by Sutter Hill Ventures, the firm behind early bets on NVIDIA, Snowflake and Astera Labs. Participants included John Doerr, National Grid Partners, Future Energy Ventures, Emerson Collective and Stanford University, following $13.5 million in seed funding raised in 2025.
GridCARE's bet is that software, not just steel, is the fastest path to powering AI growth, and that grids hold far more usable headroom than conventional planning assumes.