...
NVIDIA H200 shipments delayed to Q3  · BREAKING: Microsoft confirms 3GW data centre expansion in Asia-Pacific ·  AWS announces new sovereign cloud regions in India and UAE  · Arm-based servers now 24% of hyperscale deployments ·  EU AI Act enforcement enters phase two  · Global data centre investment hits $612B in 2026 ·  TSMC Arizona yields improve to 68% on 3nm process  · OpenAI valuation reaches $400B after latest funding round ·  NVIDIA H200 shipments delayed to Q3  · BREAKING: Microsoft confirms 3GW data centre expansion in Asia-Pacific ·  AWS announces new sovereign cloud regions in India and UAE  · Arm-based servers now 24% of hyperscale deployments ·  EU AI Act enforcement enters phase two  · Global data centre investment hits $612B in 2026
NVIDIA H200 shipments delayed to Q3  · BREAKING: Microsoft confirms 3GW data centre expansion in Asia-Pacific ·  AWS announces new sovereign cloud regions in India and UAE  · Arm-based servers now 24% of hyperscale deployments ·  EU AI Act enforcement enters phase two  · Global data centre investment hits $612B in 2026 ·  TSMC Arizona yields improve to 68% on 3nm process  · OpenAI valuation reaches $400B after latest funding round ·  NVIDIA H200 shipments delayed to Q3  · BREAKING: Microsoft confirms 3GW data centre expansion in Asia-Pacific ·  AWS announces new sovereign cloud regions in India and UAE  · Arm-based servers now 24% of hyperscale deployments ·  EU AI Act enforcement enters phase two  · Global data centre investment hits $612B in 2026

California’s 330 MW Fight Questions Compute’s Breakneck Expansion

A 330-megawatt compute facility is difficult to describe as merely a building with servers inside it. Its computing capacity depends

Share
California 330 MW

A 330-megawatt compute facility is difficult to describe as merely a building with servers inside it. Its computing capacity depends on an electrical system capable of delivering enormous power, backup generation that can sustain operations, batteries that can stabilize or support that supply, and cooling and water infrastructure that keeps the equipment operational. That physical reality has now collided with a regulatory question in Imperial County, California.

A tentative ruling from Imperial County Superior Court Judge L. Brooks Anderholt found that county officials could not treat the proposed Imperial Valley Computer Manufacturing project as a collection of largely mechanical approvals while relying on an older environmental review. The ruling calls for the county to set aside key approvals and conduct a full environmental review under the California Environmental Quality Act, or CEQA. The important issue is not simply whether this particular project moves forward. The more consequential question concerns how the industry defines a data center when the infrastructure surrounding the computing floor makes the computing floor possible.

The 330 MW Number Is Only the Beginning

The proposed project illustrates why that distinction matters. The development covers about 75 acres and includes an approximately 950,000-square-foot data center with a 330 MW load. But the project also includes a 330 MW substation, an 862 megawatt-hour battery energy storage system, cooling and water treatment facilities, four 500,000-gallon water tanks, a six-acre retention basin and a 330 MW emergency generation facility equipped with 132 natural gas generators.  Those components do not operate as independent businesses that happen to sit near one another. They form an infrastructure chain.

The substation exists to serve the computing operation. The generators provide backup capability for the same operation. The batteries support its electrical architecture. Cooling systems protect the computing equipment from heat. Water infrastructure supports those systems. The physical development therefore creates a useful test of whether environmental and permitting frameworks should examine each component separately or evaluate the infrastructure as a single operational system. The court’s reasoning points toward the latter approach. It found that grading, electrical facilities, water systems, lot changes, road changes and related infrastructure serve one cohesive construction and operation of the data center. That is a more significant proposition for the compute industry than another dispute over a permit.

Hyperscale Projects Have Become Infrastructure Packages

The data center industry has spent years making development faster by dividing complex projects into specialized workstreams. Utilities handle grid connections. Developers negotiate land and entitlements. Energy specialists arrange generation. Battery providers supply storage. Mechanical contractors build cooling systems. Water specialists handle treatment and supply. The resulting project can look like a set of parallel infrastructure transactions. Operationally, however, the separation becomes harder to defend as facilities grow.

California’s Energy Commission already recognizes that data centers can involve substantial supporting energy infrastructure. The agency notes that most data centers rely on the electric grid as their primary power source but commonly require on-site generation for backup. It also regulates certain thermal generating facilities of 50 MW or more and uses the Small Power Plant Exemption process for qualifying facilities up to 100 MW. The Imperial County dispute pushes that concept into a broader question: Can permitting frameworks continue to treat supporting infrastructure as peripheral when the compute facility cannot function without it? At 330 MW, the answer becomes increasingly difficult to separate from the physical facts.

The Bigger Risk May Be Definition, Not Delay

A full environmental review adds another regulatory step to a project and, in this case, the developer has raised concerns that the process could affect the project’s timeline.  That consequence matters in a market where developers compete for power capacity, construction windows and access to increasingly constrained infrastructure. But the more interesting consequence could involve project definition. If regulators increasingly evaluate the compute facility and its supporting infrastructure together, developers may need to establish the environmental footprint of a project earlier and more comprehensively. A power system that once appeared as a separate infrastructure decision could become part of the same environmental analysis. The same could happen with water systems, storage and on-site generation.

That does not necessarily mean every hyperscale development will face the same outcome. The California case turns on its specific facts, approvals and regulatory record. The court also found substantial evidence supporting potential environmental effects involving energy and electrical infrastructure, water supply and wastewater, air quality, greenhouse-gas emissions, hazardous materials, fire safety, traffic, public services and cumulative effects.  Still, the reasoning gives developers and local authorities another project-specific example to consider when structuring and reviewing large compute developments.  The question shifts from What permits does the building require? to What physical system is actually being approved?

Compute Capacity Could Become an Environmental Unit

That shift could prove particularly important as AI infrastructure projects continue to require substantial amounts of electrical capacity. A megawatt figure describes more than the electrical demand of servers. It can determine the scale of substations, transmission connections, backup generation, battery storage and cooling systems required to deliver reliable compute capacity. That makes power capacity increasingly inseparable from physical development.

The California Energy Commission’s own planning work reflects the growing importance of data center demand to the state’s energy outlook. The commission works with utilities to assess the likelihood that proposed projects will actually reach completion and incorporates that information into its broader electricity demand forecasting. The Imperial County case adds another layer: If a project consumes hundreds of megawatts and requires an extensive private energy architecture to support that consumption, the environmental footprint cannot easily stop at the walls of the computing building. This could eventually influence how developers structure applications, how counties evaluate infrastructure and how investors assess development risk.

The Industry May Need to Stop Thinking in Components

The most disruptive part of the ruling is therefore not that California may slow one data center. It is that the decision challenges a convenient mental model. The traditional development model can treat the data center as the central asset and everything else as supporting infrastructure. But AI compute increasingly reverse that relationship. The computing operation can drive the electrical architecture, which in turn can shape generation, storage and cooling requirements, while the resulting infrastructure can affect a project’s water and land requirements. The components remain legally and commercially distinct in many cases. They are not necessarily operationally distinct.

If environmental review follows the operational reality rather than the contractual structure of a project, developers may have less room to present a hyperscale facility as a collection of individually manageable approvals. The project could instead emerge as one infrastructure system whose impacts become visible only when the pieces appear together. That would not necessarily stop the expansion of AI infrastructure. It could, however, force the industry to define what it is actually building before it asks regulators to approve it. And that may prove to be the more consequential question raised by California’s 330 MW fight.

[simple-author-box]

More from AI Infrastructure

A 500-megawatt AI facility can represent a substantial engineering undertaking, but its resilience depends

A data center can have substantial contracted or announced power capacity while its actual

The most revealing question about putting AI infrastructure in orbit is not whether it

COMPUTE WEEKLY

The briefing that 40,000+ tech leaders read every Monday. Sharp, fast, essential.

Great! We’ve received your information.

Building an AI Startup Without Owning GPUs

Not owning GPUs has become the default, deliberate strategy for building an AI company — not a compromise founders accept reluctantly. H100 rental rates fell 64-75% in fifteen months, a dense ecosystem of neoclouds and inference-as-a-service providers now lets startups skip infrastructure entirely, and credit programs can fund a company’s first year before a founder writes a check
Most Read

A data center master plan can establish a defined technical basis before all future

A transformer can leave a refurbishment shop looking almost indistinguishable from a new unit,

Why Samsung Is Taking AI Infrastructure Offshore AI infrastructure now faces a practical challenge

AI infrastructure decisions for high-density deployments increasingly involve what happens after electricity enters the

Demand is broadening across enterprise workloads APAC’s infrastructure story is changing in ways that

Disruptor Spotlight

Cerebras Systems

The chip that makes Nvidia nervous. Cerebras’ Wafer Scale Engine is rewriting the rules of AI inference at scale.
Faster
0 x
YoY Revenue
0 x
Transistors
0 T
Market Pulse
MSFT
+1.02%
NVDA
+0.66%
AMZN
-0.078%
AMD
-6.95%
TSMC
-2.98%
Indicative only · Not financial advice
Upcoming Events
SEP
The AI Infrastructure Race (India)
WEBINAR · ONLINE
The AI Infrastructure Race: Won on Power, Land and Trust — Not Capital
MAY
0
AI Infrastructure Summit
DUBAI · IN PERSON
MEA’s premier AI infrastructure event.
JUN
0 0
Compute Forecast Summit
SINGAPORE · IN PERSON
Our flagship APAC event. Early bird open.
Latest Moves
Live
ecolab
Ecolab Deepens Cooling Strategy With $4.75B CoolIT Acquisition
Ecolab is making one of its biggest moves yet into AI infrastructure after completing its $4.75 billion acquisition of liquid cooling specialist CoolIT Systems
Pure DC AVK Europe data center microgrid Dublin 110MW AI infrastructure Ireland 2026
Pure DC and AVK Deploy Europe’s First 110 MW Data Center Microgrid in Dublin
The Pure DC Dublin microgrid has made history as Europe’s first large-scale on-site data center microgrid, launched in partnership with power solutions provider AVK at Pure DC’s campus in Ireland.
Pace Digitek
Pace Digitek Partners With MEGMEET to Expand AI Data Center Power Business
India’s AI infrastructure ecosystem continues to mature as domestic technology manufacturers move beyond traditional telecommunications and industrial markets toward high-growth digital infrastructure opportunities
Follow Compute Forecast
11K followers
1200 followers
Companies to Watch
CW
CoreWeave
Neo Cloud · $19B · IPO Watch
CB
Cerebras Systems
AI Hardware · $4.25B · Pre-IPO
G42
G42
Sovereign AI · Abu Dhabi
H
Humain
Saudi AI · $40B Fund
Latest Podcast
AI Capex, Cloud Margins & the Nuclear Bet
48 MIN · 25 APR 2026

California’s 330 MW Fight Questions Compute’s Breakneck Expansion

A 330-megawatt compute facility is difficult to describe as merely a building with servers inside it. Its computing capacity depends

Share
California 330 MW
1
847 SHARES

0
SHARES

[simple-author-box]

More from AI Infrastructure

A data center master plan can establish a defined technical basis before all future

A transformer can leave a refurbishment shop looking almost indistinguishable from a new unit,

Why Samsung Is Taking AI Infrastructure Offshore AI infrastructure now faces a practical challenge

AI infrastructure decisions for high-density deployments increasingly involve what happens after electricity enters the

COMPUTE WEEKLY

The briefing that 40,000+ tech leaders read every Monday. Sharp, fast, essential.

Great! We’ve received your information.

Global AI Infrastructure Outlook 2026

The briefing that 40,000+ tech leaders read every Monday. Sharp, fast, essential.
Download Free
Most Read

A data center master plan can establish a defined technical basis before all future

A transformer can leave a refurbishment shop looking almost indistinguishable from a new unit,

Why Samsung Is Taking AI Infrastructure Offshore AI infrastructure now faces a practical challenge

AI infrastructure decisions for high-density deployments increasingly involve what happens after electricity enters the

Demand is broadening across enterprise workloads APAC’s infrastructure story is changing in ways that

Disruptor Spotlight

Cerebras Systems

The chip that makes Nvidia nervous. Cerebras’ Wafer Scale Engine is rewriting the rules of AI inference at scale.
Faster
0 x
YoY Revenue
0 x
Transistors
0 T
Market Pulse
NVDA
$924.60
+2.4%
MSFT
$421.30
+1.1%
AMZN
$192.80
-0.6%
NVDA
$924.60
+2.4%
NVDA
$924.60
+2.4%
Indicative only · Not financial advice
Upcoming Events
MAY
0 0
DCD Global — London
LONDON · IN PERSON
World’s largest DC event. CF is media partner.
MAY
0
AI Infrastructure Summit
DUBAI · IN PERSON
MEA’s premier AI infrastructure event.
JUN
0 0

Compute Forecast Summit

SINGAPORE · IN PERSON
Our flagship APAC event. Early bird open.
Latest Moves
  • Live
Sam Altman
OpenAI appoints new Chief Infrastructure Officer to lead $100B DC programme
27 APR · OPENAI
Sam Altman
OpenAI appoints new Chief Infrastructure Officer to lead $100B DC programme
27 APR · OPENAI
Sam Altman
OpenAI appoints new Chief Infrastructure Officer to lead $100B DC programme
27 APR · OPENAI
Follow Compute Forecast
18.4K followers
12.1K followers
9.3K subscribers
41 episodes
Companies to Watch
CW
CoreWeave
Neo Cloud · $19B · IPO Watch
CB
Cerebras Systems
AI Hardware · $4.25B · Pre-IPO
G42
G42
Sovereign AI · Abu Dhabi
CW
Humain
Saudi AI · $40B Fund
Latest Podcast
AI Capex, Cloud Margins & the Nuclear Bet
48 MIN · 25 APR 2026
Scroll to Top
Seraphinite AcceleratorOptimized by Seraphinite Accelerator
Turns on site high speed to be attractive for people and search engines.