🔍 Read the full analysis: Why Electricians Are Part Of The AI Infrastructure Story on ThorstenMeyerAI.com
Get business pricing on monitors, keyboards and dev gear
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
Recent developments point to a growing infrastructure challenge behind AI data center expansion: supplying dense, rapidly changing loads with reliable electricity. Vantage reserved more than a gigawatt of factory capacity for a solid-state transformer, while NiSource and Microsoft agreed on a long-term power deal that includes new generation and a large battery. The deals highlight demand for electrical equipment and skilled workers, but they do not establish how many jobs will be created or how costs will affect customers.
Two recent power-sector deals show how the buildout of AI data centers is expanding beyond chips and servers to transformers, batteries and electrical infrastructure. Vantage reserved more than one gigawatt of factory capacity for a solid-state transformer, while NiSource and Microsoft signed a 20-year agreement tied to about 1,100 megawatts of generation and a 600-megawatt battery. The projects underline why electricians and other power specialists are part of the AI infrastructure story, though the announcements do not specify resulting job numbers or customer costs.
Vantage’s October 21 reservation concerns factory capacity for a solid-state transformer designed to convert grid power to 800-volt direct current. The source account says the equipment is intended to serve data centers with unusually high power demands. A reservation is not, by itself, confirmation that the equipment has been installed or that a specific facility is operating with it.
Five days earlier, Indiana utility NiSource signed a 20-year power deal with Microsoft. The agreement adds about 1,100 megawatts of generation and includes a 600-megawatt battery, according to the source material. The companies’ announcements indicate a long-term arrangement for power resources, but the material does not spell out how all project costs or electricity prices will be allocated.
The technical shift is tied to the power density of AI systems. The International Energy Agency (IEA) says AI server power density rose 11-fold from 2020 to 2025 and is projected to quadruple again by 2027. NVIDIA’s proposed 800-volt design would move conversion closer to the facility perimeter and distribute DC power to racks. NVIDIA claims it could reduce copper use by 45% compared with 415-volt AC distribution and improve end-to-end efficiency by up to 5% against current 54-volt systems. Those are company engineering claims, not independently verified measurements in the supplied material.
Infographic 01 · From grid to rack
AI racks near a megawatt are outgrowing the AC power chain.
Why the industry is moving to 800-volt direct current: fewer conversions, less copper, batteries on the main path. The numbers are vendor figures with different baselines, so they cannot be added up.
Today: many conversions
Typical AC chain from the grid to the rack
- Medium-voltage AC from the grid
- Transformer steps it down to low-voltage AC
- UPS converts to DC for the batteries and back to AC
- Distribution units feed the racks
- Every server power supply converts AC to DC again
Each stage loses energy, and almost all of the loss becomes heat the cooling system must remove.
800 V DC: convert once, outside
NVIDIA reference design, full-scale production planned with Kyber racks in 2027
- 13.8 kV AC from the grid
- Converted once to 800 V DC at the edge of the site, batteries on the same DC bus
- 800 V DC busway to the racks
- Stepped down inside the rack, for 1 MW racks and beyond
Bridge for existing halls: a “sidecar” cabinet next to the rack converts AC to 800 V DC, while the old chain stays in place.
Power lost per MW delivered to the rack
kW, SMA white-paper model vendor model
What NVIDIA says 800 V DC changes
NVIDIA’s engineering claims claim
Sources: NVIDIA technical blog (20 May 2025, updated 31 Jul 2025); SMA America, sponsored content in Utility Dive (26 Oct 2026). Rack density: IEA, Key Questions on Energy and AI (Apr 2026) (AI server power density 11× from 2020 to 2025, a further 4× by 2027).
Powered by Thorsten Meyer AI
Electrical Work Behind AI Growth
High-density data centers need more than power contracts. They also require equipment and work to connect, convert, distribute and manage electricity safely, from utility connections and substations to transformers, batteries, switchgear and systems inside the facility. That makes electricians and power engineers part of the practical capacity buildout: a site cannot use its computing equipment at scale without a dependable electrical system.
The bottleneck may be time as much as supply. The IEA puts transformer delivery times at two to three years and says grid connections can take five to ten years in many places. Those delays can shape when data centers open, even when operators have financing and chips. They also make planning and installation work central to whether announced capacity becomes usable infrastructure.
The buildout has implications beyond technology companies. More data center demand can prompt utilities to add generation, storage and transmission infrastructure. Who pays for those additions—and whether costs reach household bills—depends on local rules and contract terms. The source reports US residential electricity averaging 18.21 cents per kilowatt-hour in 2026, compared with 17.30 cents in 2025, but that national figure alone does not show how much of the change is attributable to data centers.
Popular right now
Roborock RockNeo Q110H Wire-Free Robot Lawn Mower
iRobot Roomba Max 715 Pet Robot Vacuum with AutoEmpty Dock
Dyson Car+Boat Cordless Handheld Vacuum
As an affiliate, we earn on qualifying purchases.
From Server Racks to Grid Connections
Traditional data centers receive alternating-current power, then use transformers and other equipment to convert it through several stages before servers receive the low-voltage DC they use. Each conversion loses some energy, much of which becomes heat that cooling systems must remove. This arrangement served earlier racks drawing tens of kilowatts, but AI systems are pushing designs toward much higher loads.
NVIDIA’s reference design targets one-megawatt racks and proposes converting 13.8-kilovolt grid power to 800-volt DC near the edge of a facility. The company says a rack supplied at 54 volts could require up to 200 kilograms of copper busbar. It plans full-scale production of its design with Kyber systems in 2027. Those figures and projections come from NVIDIA, not a neutral field evaluation.
A Utility Dive article published October 26 describes another path: a conversion cabinet, or “sidecar,” that can let 800-volt racks operate in data center halls originally built for AC. The piece was sponsored by inverter maker SMA America, and its modeled energy-loss figures should be read as vendor-sponsored estimates. The IEA separately projects global data center electricity use rising from 485 terawatt-hours in 2025 to 950 terawatt-hours in 2030. Projections describe expected demand, not a guaranteed outcome.
“Leaving no room for compute”
— NVIDIA, in its 800 VDC technical blog
Costs, Jobs and Deployment Timelines
The announcements do not say how many electricians or other workers the projects will employ, whether hiring will be permanent, or how much work will go to local contractors. The Vantage factory-capacity reservation does not establish when equipment will be delivered or installed. The source material also does not provide the full terms of the NiSource-Microsoft agreement, including how construction costs and risks are divided.
It remains uncertain how quickly 800-volt DC systems will move from reference designs and vendor proposals into broad deployment, or how their performance will compare with existing AC systems in independent operating data. NVIDIA’s efficiency and copper figures and SMA’s modeled conversion losses are attributed claims, not independent measurements supplied here. The effect of data center expansion on household electricity bills will vary by region and depends on utility investment, regulation, contracts and electricity demand from other sources.
From Reservations to Working Systems
The next evidence will come as the announced power resources move toward construction, equipment delivery and operation. Vantage’s factory-capacity reservation may be followed by more detail on production and deployment schedules. For the NiSource-Microsoft arrangement, project milestones and public filings could clarify how the generation and battery will be developed and how the agreement fits into local power planning.
For the workforce, hiring plans, contractor awards and project schedules would show what kinds of electrical roles are needed and where. For the technology, independently reported operating data could test vendor claims about efficiency, copper use and battery performance. Until those details emerge, the confirmed development is a set of major commitments—not a complete account of their labor, cost or grid effects.
Key Questions
What are the recent developments behind this story?
Vantage reserved more than one gigawatt of factory capacity for a solid-state transformer on October 21, 2026. NiSource and Microsoft signed a 20-year power deal on October 16 that includes about 1,100 megawatts of generation and a 600-megawatt battery.
Why do AI data centers need different electrical infrastructure?
AI servers can concentrate substantial power demand in individual racks and may change their electricity use rapidly. High-voltage distribution, transformers and batteries are among the systems being considered to deliver and manage that power.
Does the 800-volt DC design have proven efficiency gains?
NVIDIA claims up to 5% better end-to-end efficiency than current 54-volt systems, alongside lower copper use. The supplied material identifies these as company figures, not independently verified results from operating facilities.
Will these projects create more electrician jobs?
They point to demand for electrical construction, installation and maintenance work, but the announcements provide no job totals or hiring schedules. Employment effects will depend on project scope, timing and how the work is contracted.
Will data center expansion raise household electricity bills?
That is not established by these deals alone. Costs depend on regional utility planning, contract terms and regulatory decisions; the national electricity-price figures cited in the source do not isolate data centers’ effect on bills.
Source: ThorstenMeyerAI.com
