Investment Insights: GridCARE

1. The Big Picture

By 2030, data centers could account for up to 60% of total U.S. electricity load growth — more than EVs, industrial electrification, and every other sector combined1. Power demand from data centers alone is expected to more than double, from roughly 60 GW today to as much as 134 GW2. Globally, data center electricity consumption could reach between 980 TWh and 1,600 TWh by 2030, nearly double current levels. The electricity grid is facing its most significant stress test in a generation.

Against this backdrop, access to power has become the binding constraint. Interconnection queues have swelled to extraordinary levels: ERCOT’s large load interconnection queue reached 410 GW in March 2026, up from just 63 GW a year prior3. More than 70% of those requests are coming from data centers4 — a surge that has, in ERCOT’s own words, „outgrown the process established for reviewing these large loads.“ The pattern is not unique to Texas. Utilities across the country continue to rely on conservative planning processes that systematically underutilize existing capacity, while developers wait years for interconnection studies that may never convert to actual connections.

The consequences are not abstract. Developers unable to secure grid connections are turning to on-site fossil generation as a bridge, locking in avoidable emissions at precisely the moment when the energy system needs to decarbonize. Solving the „speed to power“ problem is not just a commercial imperative — it is a prerequisite for a credible clean energy transition.

2. Why It Matters

Utilities have historically planned for peak demand using conservative assumptions that leave significant latent capacity stranded. Analysis from Stanford University suggests that system load remains between 30-40% of peak capacity under normal operating conditions across major U.S. balancing authorities.5 That unutilized headroom represents an enormous opportunity — if only the tools existed to identify and operationalize it safely.

The challenge is not simply technical. Interconnection is a deeply regulated, relationship-driven process involving utilities, regulators, developers, and grid operators. Most attempts to address the bottleneck have focused on either capital-intensive transmission upgrades or software tools that optimize existing assets — without touching the upstream access problem that keeps developers in queue for years. The result is a market that has moved fast on ambition but slow on execution, leaving gigawatts of potential capacity stranded and developers waiting.

3. What They Do

GridCARE is the defining platform in Power Acceleration — a new category that addresses all three things a data center developer needs to do with power: find it, activate it, and operate it. Built on Stanford-developed technology and led by Amit Narayan, the serial entrepreneur behind AutoGrid (acquired by Schneider Electric), GridCARE takes a fundamentally different approach to the interconnection problem.

Energize™, GridCARE’s core AI platform, delivers three integrated capabilities for end-to-end data center energy activation and delivery. Power Finder surfaces validated capacity opportunities against proprietary utility network models. Power Activation goes inside the utility’s planning process with the utility’s own model, using physics-based AI to unlock capacity through flexibility that the traditional queue cannot see. Power Operations manages the resulting flexible interconnection agreement in real time after energization, dispatching flexibility and maintaining contracted obligations continuously. GridCARE’s business model is tied to the success of their partners.

Also, GridCARE’s engagement model creates compounding value for both utilities and data centers: each utility partnership drives multiple downstream data center opportunities within a given territory.

GridCARE’s founding team combines exceptional depth across grid software, AI applications for energy systems, and utility engagement. Amit Narayan (CEO) brings a track record of building and scaling utility-grade AI platforms. Co-founder and CTO Ram Rajagopal, Professor of Electrical Engineering at Stanford, and co-founder Liang Min, Managing Director of the Bits & Watts Initiative at Stanford, and Arun Majumdar, Dean at Stanford Doerr School of Sustainability provide the scientific and regulatory credibility that is essential in a conservative, trust-driven market. The team is further supported by an advisory board spanning utilities, regulators, and hyperscaler energy leaders.

„The interconnection process wasn’t designed for the pace of demand the grid is now facing. We built GridCare to give utilities and developers a shared AI-driven approach for moving faster without compromising reliability.“ — Amit Narayan, CEO, GridCARE

4. How It’s Going

GridCare secured its flagship utility partnership with Portland General Electric in mid-2025, unlocking 400 MW of data center capacity that would otherwise face multi-year queues under conventional interconnection processes. The PGE partnership demonstrates the platform’s core commercial thesis: that utilities can offer faster, more certain connection timelines to data center developers, unlocking value on both sides of the relationship. A subsequent partnership with National Grid extends this model to the Northeast, adding a second major territory and broadening the pipeline of downstream data center opportunities.

The Series A round of $64 million, led by Sutter Hill Ventures, is a meaningful signal of market conviction. Sutter Hill — whose portfolio includes NVIDIA, Snowflake, and Palo Alto Networks — brings a concentrated, hands-on investment model that serves as strong validation of GridCARE’s potential to become the foundational energy layer for AI infrastructure. National Grid’s participation as a strategic investor further reinforces both the commercial and regulatory credibility of the platform.

“Utility interconnection queues have become the hidden bottleneck of the AI buildout. GridCare is the first platform we’ve seen that addresses the problem at the right layer — grid physics, operational data, and commercial structure, all in one place.” — Patrick Elftmann, Managing Partner at Future Energy Ventures

5. Looking Ahead

Future Energy Ventures (FEV) is pleased to be participating in GridCare’s Series A alongside Sutter Hill Ventures, Doerr Capital, National Grid Partners, and other leading investors.

The investment will strengthen GridCARE’s go-to-market and accelerate product development as the company scales its platform across additional utility territories to meet immense customer demand. Over the next 12 to 24 months, the measure of success will be straightforward: new utility territories activated, gigawatts contracted, and a data advantage that widens with each partnership.

Co-authored by Anna Trendewicz and Patrick Elftmann

 

Sources

  • BCG, Breaking Barriers to Data Center Growth, October 2025 — https://www.bcg.com/publications/2025/breaking-barriers-data-center-growth
  • S&P Global, Data center grid-power demand to rise 22% in 2025, nearly triple by 2030, October 2025 — https://www.spglobal.com/energy/en/news-research/latest-news/electric-power/101425-data-center-grid-power-demand-to-rise-22-in-2025-nearly-triple-by-2030
  • RTO Insider, ERCOT Large Load Interconnection Queue Hits 410 GW, March 2026 — https://www.rtoinsider.com/122627-rtc-deployed-ercot-takes-on-new-challenges/
  • Utility Dive, ERCOT’s large load queue jumped almost 300% last year, January 2026 — https://www.utilitydive.com/news/ercots-large-load-queue-jumped-almost-300-last-year-official/808820/
  • U.S. Transmission System Utilization Study Phase 1: WECC, R. Prabha, L. Min, Bits and Watts Initiative, Stanford University, 2025, https://nicholasinstitute.duke.edu/articles/study-finds-headroom-on-grid-data-centers

Discover more