There is no shortage of power plants waiting to be built. There is a shortage of ways to connect them. That distinction is the whole story.
The debate over whether the U.S. can power AI usually gets framed as a generation problem — do we have enough electricity? The data says the real constraint sits one layer down, in the interconnection queue: the multi-year line every new power plant and battery must wait in before it can plug into the grid. Roughly 2,061 GW of capacity is queued today — far more than enough — but most of it will never get built, and what does will arrive years after the demand. Here is what the primary sources actually show.
Every credible forecaster agrees U.S. electricity demand is climbing after two flat decades, and that data centers are the driver. The EIA calls it the strongest four-year demand growth since 2000 — the first four straight years of growth since 2007 — projecting +1% in 2026 and +3% in 2027. Data centers used about 192 TWh (4.7% of U.S. electricity) in 2024, up 14% in a single year, according to Berkeley Lab's June 2026 update.
Until recently, forecasters disagreed sharply on how high, how fast. The 2026 editions have closed ranks: LBNL now projects 11.8% by 2030 (range 9.5–15.3%), squarely inside EPRI's 9–17%. The forecasts are converging — upward, toward roughly 12% of U.S. electricity by 2030. That convergence is the honest headline.
Here is the counterintuitive core. As of end-2025, about 2,061 GW of generation and storage — some 8,244 projects, roughly 1.5× the entire installed U.S. fleet — was seeking to connect to the U.S. grid. That is multiples of what the buildout needs. The problem is throughput: the median time from interconnection request to switched-on has nearly tripled, from 22 months for projects built in 2008 to 61 months — just over five years — for those built in 2025.
And the wait is only half of it. Most queued capacity is never built at all. Of everything that requested interconnection between 2000 and 2020, only 13% was operating by end-2025 — while ~75% was withdrawn. Even a signed interconnection agreement is no guarantee: 41% of capacity that signed one (2000–22) later withdrew. The queue is an attrition filter, not a build pipeline.
The queue has now shrunk two years running — total volume fell about 10% in 2025 as more than 750 GW withdrew, with solar (−19%), storage (−16%) and wind (−19%) all down. One resource ran the other way: natural gas surged 86% to 253 GW. The queue is still mostly clean by volume — gas is about an eighth of it — but gas is the only major category growing — an early signal that developers are chasing firm, always-on power they can count on connecting.
On the supply side, the IEA's global outlook (to 2035) has renewables covering roughly half of data-center demand growth, with natural gas (notably in the U.S., ~+175 TWh) and nuclear each adding about the same, and the first small modular reactors arriving around 2030. But every one of those has a lead-time problem:
The fixes that need no new steel are the ones that work fastest: grid-enhancing technologies and advanced-conductor reconductoring can lift the capacity of lines already in the ground (per the U.S. DOE and GridLab), and making large new loads flexible — willing to curtail briefly at peaks — can free meaningful headroom on the existing grid. Those are the levers with the shortest path to a connected megawatt. And where the wires can't come fast enough, the hyperscalers are routing around the queue entirely — on-site gas, nuclear restarts and SMRs — which is the subject of our companion analysis, powering around the queue.
National averages hide the real risk, because the load concentrates. About half of U.S. data centers under development sit in pre-existing large clusters (IEA), stacking new demand onto grids already carrying it. Virginia is the extreme case — the only state where data centers already use about 25% of its electricity, a share that could reach 39–57% by 2030 (EPRI) — and on EPRI's medium scenario, seven more states could cross 20% by decade's end. That geography is exactly what the regional maps track, market by market.
Not power plants — the grid interconnection queue. The median wait to connect a project has more than doubled to about 5 years, and around 75% of proposed capacity is ultimately withdrawn. Compute scales in months; interconnection scales in years.
The median interconnection wait is about 5 years, and building new transmission takes 4–8 years. The IEA estimates ~20% of planned data-center projects risk delay from grid constraints.
Generation isn't the binding constraint — throughput is. The queue is still mostly clean by volume — gas is about an eighth of it; the question is whether the right wires, in the right places, get built before the GPUs show up.
Natural gas — up 86% to 253 GW in 2025 — as developers chase firm, always-on power they can count on connecting, even as solar, storage and wind requests fell.
The full picture, mapped. The Gridlas report puts these forecasts, the queue data, and five regional deep-dives onto the grid — with high-res maps and the underlying dataset (CSV/GeoJSON), built entirely from public data.
Get the report →Prefer the free 3-page summary first? The 8 findings and the chart that explains the AI power crunch — straight to your inbox.
No spam. Unsubscribe anytime.