Both sides of the data center debate: fourteen arguments, with the counters

Fourteen arguments, both directions

Each argument below is a complete unit: the claim, the evidence, what the other side says back, and the reply. Read both sides and you have seen the whole battlefield from both hills. Neither list is ours. Both are built only from arguments real advocates actually make.

Handbook text revised September 16, 2026. Every argument is checked against new evidence daily; the date beside each one is the last time something changed for it. Last daily check: 2026-09-16T10:30:22.

Part 5

The four points where the sides genuinely collide

Most of this debate is two sides answering different questions. These four are real disagreements, and a serious exchange should end up here.

Did families actually pay more? One study finds no measurable subsidy in what customers were billed. The grid watchdog finds billions in costs caused by data center demand.⁸ ¹⁷ These are not flatly contradictory, because who caused a cost and who got billed for it are different questions. Whether that difference ever reaches a family's bill is genuinely unresolved.
Is flexibility real or just promised? The 100 gigawatts of room depends on data centers easing off during peak hours.¹¹ Live tests are running at nine sites.¹² Most contracts signed today do not require it.¹¹
Do construction benefits count? Both major studies find the benefits concentrated in the building phase.⁶ ²² Construction unions think that is the whole point.⁷ Opponents think that is the whole problem.
Who eats the loss if a project fails? Contracts requiring minimum payment push that risk onto the data center.⁹ Whether those terms become standard everywhere, and whether they survive the first big bankruptcy, nobody knows yet.