001 / CASE FILE / PROJECT CAMELLIA

What We Know—and Still Don’t Know—About Project Camellia

A source-traceable Georgia investigation separating verified facts, announced claims, estimates, and unanswered infrastructure questions.

FLAGSHIP EVIDENCE BRIEF

Prepublication report v0.1Last verified August 26, 202613 pages · 20 linked sources

THE BOTTOM LINE

Large potential benefits. Unusually large resource demands. Important protections that still need measurable terms.

Project Camellia is a proposed AI infrastructure campus with a public record strong enough to understand its announced scale—but not yet strong enough to verify several of its most important outcomes. This report explains what is established, what is projected, and what remains unknown without arguing for automatic approval or automatic rejection.

4.4Msquare feetFiled physical program
3,210megawattsPeak connected-load estimate
30,000gallons/dayFiled water-demand estimate
65%imperviousFiled planning estimate

FIVE STRONGEST CONCLUSIONS

What the evidence supports today.

Each conclusion is deliberately bounded. A filed estimate proves that a figure was submitted—not that the predicted result has occurred.

01

The scale is established

Filed state records describe four buildings totaling 4.4 million square feet and a 3,210 MW peak connected load.

02

Review is not final approval

The DRI process is advancing, but “Certified Complete” does not mean every construction, utility, or environmental approval has been issued.

03

Closed-loop cooling is promising—not water-free

The design may materially reduce ongoing withdrawals, but the final agreement, initial fill, peak demand, drought rules, meters, and audits are not yet public.

04

Ratepayer safeguards exist

Large-load rules, long-term contracting, financial security, and flexible demand can protect other customers. Project-specific operating terms remain only partly visible.

05

The fiscal case needs reconciliation

One-time land proceeds, recurring taxes, projected sales-tax revenue, abatements, public costs, and community benefits should be shown in one year-by-year model.

WHAT STILL NEEDS PROOF

Five systems under continuing review.

An unanswered question is not evidence of wrongdoing. It is a research item that should be resolved with records, measurable terms, and accountable decisions.

01

Project status

Which elements are proposed, approved, contracted, under construction, or operating?

02

Electricity

What capacity is requested, how will service be delivered, and which costs or risks remain unresolved?

03

Water + cooling

What cooling design and water source are described, and which figures are measured rather than estimated?

04

Land + permitting

Which zoning, stormwater, wetlands, traffic, safety, and construction approvals apply?

05

Public value

How do announced investment, employment, incentives, taxes, and local obligations compare with the evidence?

OUR STANDARD

Conditional confidence, backed by public evidence.

The project should be judged by whether it converts extraordinary scale into durable public value while measurably limiting resource use and environmental harm. Claims should earn trust through contracts, permits, operating data, independent verification, and consequences when commitments are missed.

Read the methodology and sources

CORRECTIONS + EVIDENCE

Challenge the record constructively.

Source submissions, documented corrections, and substantive responses are welcome. Material changes will be dated and described. Do not send confidential or legally protected material without arranging a secure channel first.