For entrepreneurs with a provocative vision for energy & IT; NGP is the only utility venture & innovation fund that taps into the global network of National Grid, accelerating companies with investment, innovation, incubation & biz dev; realizing the full potential of energy.

National Grid Partners

As data centers scale in size from 20 MW to enormous 2GW interconnection requests, the once-steady rhythm of the electric grid is being rewritten by the pulse of AI. At the recent VERGE climate-tech conference in Silicon Valley, industry leaders discussed how data center growth is rewriting the rules of grid reliability. 

A roundtable led by Aaron Panzer, Head of Venture Acceleration at National Grid Partners, explored how utilities, hyperscalers, and innovators can work together to balance speed, cost, and reliability in a grid that’s straining to keep up. 

Insights from VERGE 2025

Utilities and their customers are re-examining the traditional “five-nines” standard for network uptime. With AI workloads multiplying and power supplies tightening, reliability now means more than just keeping the lights on: 

Deepa Poduval, SVP of Global Sustainability at engineering giant Black & Veatch, reminded the group that interruptions are now more expensive than electricity bills for many clients. Mission-critical facilities—from data centers to defense—are prioritizing uninterrupted operations and investing in hybrid microgrids blending solar, landfill gas, and diesel for resilience. 

Aroon Vijaykar, SVP of Strategy and Commercial at startup Emerald AI, painted a vivid picture: “The unstoppable force of AI meets the immovable object of the utility grid.” Emerald’s software-only approach orchestrates artificial intelligence workloads to shape load dynamically, turning computing demand into a flexibility asset. 

Savannah Goodman, Head of Google’s Advanced Energy Labs, said her team likewise is re-thinking the sacred “five-nines” uptime model. In today’s power-constrained world, she said, reliability is becoming situational, depending on workload. Data centers, Goodman noted, “need to be responsible grid citizens.” 

Amaani Hamid, Head of Flexibility as a Service at Octopus Energy, flipped the industry narrative, arguing reliability isn’t a constraint—it’s an opportunity. Aggregating distributed energy resources (DERs) into virtual power plants, she said, can deliver grid support within minutes. 

Reliability Redefined

Panelists agreed innovation is happening fast, from load orchestration to demand response and “CapEx as a Service” models. The goal: speeding interconnections, cutting costs, and building smarter flexibility into every layer of the grid. 

Google’s demand-response contracts with TVA, Entergy Arkansas, and Indiana Michigan Power link data center capacity to real-time grid needs. Load-shifting and behind-the-meter storage (such as batteries, generators, and even software-defined assets) are now part of the playbook. 

Emerald AI, partnering with NVIDIA, is developing flexible data center reference architectures that could unlock faster interconnections and larger power budgets without overloading local grids. 

Octopus Energy’s Kraken platform is pioneering “CapEx as a Service,” helping corporate customers monetize their flexibility by aggregating idle assets and optimizing energy use. 

Participants also highlighted the growing popularity of time-of-use rates, clean-transition tariffs, and “bring-your-own-capacity” models. These new market tools are designed to share costs fairly while speeding new connections. 

Emerging Solutions in Motion

And even as technology quickly morphs, the regulatory and market landscapes also are shifting. Federal and state leaders are accelerating interconnection timelines and encouraging flexibility for large energy loads.

The Federal Energy Regulatory Commission, for instance, has called for accelerated interconnection for flexible large-energy—a top-down push for adaptability across public utility commissions.

Yet gaps remain: While FERC governs wholesale markets, distribution-level flexibility still lacks standardized incentives.  Panelist noted this “fifty-state patchwork” of regulation complicates investment. 

Regulation Meets Competition 

In the age of AI and electrification, reliability is being re-engineered. The industry is moving from “always-on” to “smart-on”—where flexibility, automation, and collaboration define success. 

Speed > Cost. For hyperscale data center operators, the race to capacity is now the top driver. Reliability, affordability, and carbon intensity form a new, three-way balancing act. 

Flexibility is the new redundancy. Behind-the-meter storage, demand-response programs, and smarter building-management systems are the fastest entry points for corporates entering grid flexibility markets. 

Social license matters. Communities expect data centers to give back—through local resilience, shared benefits, and transparent engagement. 

Partnerships power progress. From utility co-planning to corporate transparency on load growth, collaboration beats isolation in solving the grid-AI paradox. 

In the AI era, innovation and responsibility must move in lockstep. Whether you’re a utility, an innovator, or an energy-intensive corporate, the next step is clear: Build, pilot, and scale solutions that make “smart-on” reliability the new standard for a clean, resilient grid.

Takeaways: The New Reliability Playbook

By: Marit Lyngbaek, National Grid Partners

Beyond Backup: How Corporations Can Keep The Grid Online