AI is rewriting the energy equation faster than the grid can keep up. Data centers already account for roughly half of all new U.S. electricity demand growth, and the trajectory is steep – projections suggest data center power consumption could double or triple by the end of the decade. That demand has to come from somewhere, and increasingly, it’s putting direct pressure on the grid infrastructure that delivers it.
On April 20, 2026, the White House responded by invoking Section 303 of the Defense Production Act – a wartime-era authority – to accelerate domestic production of critical grid components. Five separate determinations cover grid infrastructure equipment, natural gas, LNG, petroleum, and large-scale energy systems. While the budget will be put before Congress later this year, the Department of Energy now has the authority to deploy loans, purchase commitments, and cost-sharing arrangements to expand domestic manufacturing capacity for the transformers, switchgear, and other components the grid desperately needs.
The reasons are straightforward. The supply chain can’t keep pace – and much of it isn’t domestic. Lead times for large power transformers have stretched to 128 to 144 weeks, and roughly 80% of large power transformers installed in the U.S. are imported, primarily from countries whose trade relationships with the U.S. are increasingly uncertain. The DPA determinations explicitly cite the risk of depending on foreign-sourced components for infrastructure the government now considers a matter of national security. Reshoring that supply chain is a multi-year effort, and in the meantime, utilities need to know exactly what’s deployed, what’s aging, and what’s at risk – because replacing a failed transformer you didn’t know you depended on is a very different problem than replacing one you’ve been tracking.
What This Means for Utility Networks
DPA-driven federal investment will come with reporting, documentation, and auditability requirements. With them, IT teams will need a clear view of what’s in a network, how infrastructure changes over time, and how operational reality differs from the intended state of the infrastructure; exactly the kind of ground truth a utility needs when applying for DOE grants, justifying capital projects, or demonstrating supply chain readiness to federal auditors.
Another key factor that the DPA is meant to address is speed to deployment. Craig Watson, OT Solutions Architect at World Wide Technology (WWT), sees an increase in utility customers trying to modernize their telecommunication networks, scale automation, and deliver consistent capabilities across complex enterprise environments. For utilities expanding OT/IT networks (substations, private LTE, SCADA), this matters enormously.
The DPA invocation is not just a supply chain story — it’s a network readiness story. Utilities that can demonstrate:
● A verified, real-time system of record for all network assets (IT and OT)
● Automation pipelines that can scale with rapid infrastructure deployment
● Governance and auditability for federal compliance
will be the best positioned to access DOE financing and execute DPA-aligned projects on time.
Power utilities that have a combination of an architecture-led delivery capability and a foundational data model and automation layer that makes it all governable and auditable will be well positioned to take advantage of this unprecedented event and capitalize on providing a well-poised and resilient, hardened grid infrastructure.
Speed to Scale Requires Knowing What You Have
The DPA invocation signals that the federal government views grid modernization as an issue of national urgency. But urgency without operational readiness is just pressure. And the single biggest barrier to moving fast on infrastructure – whether you’re integrating new generation sources, upgrading substation equipment, or expanding distribution networks – is not having an accurate, authoritative picture of what’s already deployed.
Most utility infrastructure teams know this intuitively. They’ve lived with the gap between what their systems say exists and what’s actually in the field. Configuration management databases built for IT ticketing workflows don’t capture the network-level detail that operations teams need – interface configurations, circuit paths, rack elevations, IP assignments, cabling. The result is that every change, every upgrade, every incident response starts with someone manually reconciling what they think is true against what they can verify.
That friction is manageable when you’re operating at a steady state. It becomes untenable when you need to scale fast.
A Fortune 100 utility we work with discovered it after years of relying on a general-purpose ITSM platform as their de facto infrastructure record – they found it simply couldn’t serve as a network source of truth. The data wasn’t granular enough. The reconciliation process was manual and error-prone. They had no reliable way to validate what was deployed against what was intended. They’re now rebuilding their approach around a purpose-built infrastructure documentation platform – one that treats network and OT assets as first-class objects, not afterthoughts in a service catalog.
OT and IT Are Converging – Whether You’re Ready or Not
The DPA determinations don’t distinguish between IT and OT infrastructure. They cover grid equipment broadly – the physical systems that generate, transmit, and distribute power. But on the ground, the teams managing these systems are increasingly being asked to treat them as a unified environment.
Substations are getting IP-connected. SCADA networks are being modernized. Field devices that were once isolated are now part of monitored, managed networks. And the tooling that worked when OT was air-gapped and managed through spreadsheets and site visits doesn’t work when those same assets need to be inventoried, monitored, and automated at the same cadence as enterprise IT.
One of our customers – an energy company that quadrupled in size through acquisitions – learned this the hard way. They inherited two entirely separate networks – a corporate IT environment and an OT/SCADA environment – with overlapping IP space and no shared documentation. Every acquisition compounded the problem. Their team described automated discovery and drift detection as “a must” for integrating acquired infrastructure without introducing operational risk.
A major North American pipeline operator is going through a similar transition, with a systems integrator leading a formal OT/IT asset migration – bringing operational technology assets into the same documentation and management framework as their enterprise network. The driver isn’t theoretical convergence. It’s that operating two parallel, disconnected inventory systems creates blind spots that become dangerous as infrastructure scales.
Modernizing the NOC for the Grid of the Future
The utilities that are best positioned to respond to the DPA’s urgency are the ones that have already invested in modernizing how they operate their networks – not just what equipment they deploy.
At a large investor-owned electric utility in the western U.S., a network operations center modernization initiative replaced fragmented, unreliable infrastructure records with a single authoritative source of truth. The results were concrete: zero-touch provisioning with same-day service activation, per-device software upgrade times reduced from hours to minutes, and more than a thousand engineering hours saved per month. The foundation for all of it was accurate, continuously validated infrastructure data – the kind of data that makes automation reliable instead of risky.
This pattern – invest in the data foundation first, then build automation on top – is emerging as the playbook for utilities that need to move fast without introducing the configuration errors and inconsistencies that cause outages. It’s the opposite of the traditional approach, where teams
automate first and hope the data catches up. When your device count is about to increase by an order of magnitude, hope is not a strategy.
What Section 303 Actually Demands of Infrastructure Teams
The DPA invocation is, at its core, a signal that the federal government expects grid infrastructure to scale significantly and quickly. For the teams responsible for building and operating that infrastructure, the practical implications are clear:
Know what you have. You cannot modernize, expand, or secure infrastructure you haven’t accurately documented. If your inventory lives in spreadsheets, tribal knowledge, or an ITSM tool that wasn’t designed for network-level detail, that gap becomes a liability – not just operationally, but in the context of federal compliance and funding programs that will increasingly require auditable infrastructure records.
Treat OT assets like network assets. The era of managing operational technology through separate, disconnected systems is ending. Substations, field devices, SCADA controllers, and distribution equipment need to be inventoried, tracked, and validated with the same rigor as routers and switches. The organizations doing this well are using platforms purpose-built for infrastructure documentation – not general-purpose CMDBs retrofitted for OT.
Build for automation, not just visibility. Accurate infrastructure data isn’t just for dashboards and reports. It’s the foundation that makes zero-touch provisioning, configuration validation, drift detection, and automated remediation possible. As device density on the grid increases by an order of magnitude, manual operations become a bottleneck that no amount of staffing can solve.
Validate continuously. A source of truth that’s accurate on day one and stale by day thirty isn’t a source of truth – it’s a snapshot. The infrastructure data foundation needs to be continuously validated against what’s actually deployed, with automated discovery and reconciliation catching drift before it becomes an incident.
The Grid Is Critical Infrastructure. Treat Its Data That Way.
The Defense Production Act is usually invoked for manufacturing capacity – building more transformers, more switchgear, more of the physical components the grid needs. But manufacturing capacity without operational capacity is a bottleneck in a different place. You can produce transformers faster, but if the teams installing and managing them can’t keep pace – because they’re working from inaccurate data, manually reconciling inventories, and operating blind in parts of their network – the bottleneck just moves downstream.
The organizations that will scale fastest are the ones that have invested in the operational data foundation: accurate, granular, continuously validated infrastructure documentation that spans IT and OT, supports automation, and gives every team – from NOC operators to field technicians to compliance – a shared, authoritative view of what exists and what it’s supposed to look like.
AI is driving unprecedented demand on the grid. The DPA is accelerating the supply-side response. The question for infrastructure teams is whether their operational tooling can keep pace with both. For most, the honest answer is: not yet. But the path forward is clear – and it starts with knowing what you have.
Kris Beevers is the co-founder and CEO of NetBox Labs, where he is building the future of network automation and infrastructure management. An engineer by background, Kris has spent over 15 years helping organizations design, deploy, and scale some of the world’s most demanding network and infrastructure environments – from AI datacenters and neoclouds to global enterprises and industrial OT systems. He is also the former co-founder and CEO of NS1, the managed DNS provider acquired by IBM in 2023. Deeply rooted in open-source and community-driven innovation, Kris remains actively engaged with the NetBox community and is passionate about empowering engineers through practical, scalable infrastructure automation, believing that the best technology is built not just for users, but with them.
