A Queue That Outgrew Its Own Rules
For two years, Texas has been the easiest place in America to plug an AI data center into the grid, and that very openness has now become its problem. The Electric Reliability Council of Texas is tracking more than 438,000 megawatts of large-load interconnection requests, with nearly 89 percent of that volume coming from data centers alone. The queue has ballooned from 63 gigawatts in late 2024 to well over 200 gigawatts, a number that exceeds the entire peak demand of the ERCOT system several times over.
The old process treated every request as a standalone study, evaluated in the order it arrived. That made sense when a single 50 megawatt factory came along once a quarter. It collapses when hundreds of gigawatt-scale campuses arrive at once, each one assuming the grid will be there when construction finishes. We see Batch Zero as ERCOT's admission that first-come, first-served can no longer triage demand of this magnitude, and that some form of rationing was inevitable.
What the Commission Actually Approved
On June 18, 2026, the Public Utility Commission of Texas voted to approve ERCOT's Batch Zero interconnection process. The mechanism applies to loads of 75 megawatts and greater, and it replaces project-by-project review with grouped batch studies. ERCOT will assess a cohort of qualified projects together, allocate available grid capacity across them, and identify the specific transmission upgrades each batch requires before anyone is cleared to energize.
ERCOT describes this as the first ISO-level batch framework of its kind in the country, which means other grid operators facing the same data center crush will be watching how it performs. The process also builds in pathways for customers who bring their own onsite generation and for those willing to accept curtailment during transmission constraints. In effect, the operators most able to be flexible about when they draw power move to the front of the line.
The Timeline That Matters for Site Planners
The dates are where this gets real for anyone holding a Texas land position. ERCOT expects to notify applicants of their project classifications by August 2026. The first formal Batch 1 application window opens in summer 2027, and a final transmission plan is expected in fall 2027. That is a long runway for a hyperscaler that assumed grid access was a formality, and it injects genuine schedule risk into campuses that were penciled in for 2027 and 2028 energization.
For CIOs and infrastructure leads, the practical takeaway is that a signed land deal and a utility letter of intent are no longer enough to model a go-live date. The classification ERCOT assigns this summer effectively becomes a project's place in the pecking order, and projects that miss the early cohorts may wait years for transmission to catch up. The premium on shovel-ready power, including behind-the-meter generation, just went up.
The Process Behind the Process
ERCOT did not write these rules in a vacuum. Jeff Billo, the operator's vice president of interconnection and grid analysis, noted that the stakeholder effort produced more than 200 hours of live discussion, with workshops that averaged roughly 500 participants each. "Our existing process really was not designed for the volume of large load interconnection requests," Billo said, a candid acknowledgment that the system was improvised long before the AI build-out arrived.
ERCOT chief executive Pablo Vegas framed the stakes in broader terms. "Texas is experiencing an energy transformation unlike anything we have seen before," Vegas said. The volume of engagement signals that this was not a quiet regulatory tweak but a contested redesign of who gets power and when, with billions of dollars of data center investment hanging on the outcome of each batch study.
Cost Risk Moves Onto the Operator
Batch Zero does not arrive alone. On June 10, Governor Greg Abbott issued a directive requiring data centers to fully fund the electric infrastructure needed to serve them, including new power lines, substations, and grid upgrades, rather than passing those costs to residential ratepayers. Read together, the two moves redraw the economics of building in Texas: operators now bear both the timing risk of the batch queue and the capital cost of the wires that connect them.
This is a deliberate political settlement. Texas wants the jobs and tax base of the AI boom without the backlash of higher household electric bills, and Batch Zero plus the Abbott directive is how it threads that needle. For executives, the lesson generalizes well beyond ERCOT. As grids everywhere strain under AI demand, the cheap, frictionless power that made certain regions attractive is being repriced, and the cost of interconnection is becoming a line item that belongs in the business case from day one.
Why This Is a Template, Not an Exception
We expect Batch Zero to be copied. Every major grid operator is staring at the same arithmetic: interconnection queues stuffed with speculative gigawatts, real projects competing with placeholders, and transmission that takes the better part of a decade to build. A batch study process lets an operator separate serious capital from optionality, and it gives regulators a defensible way to say no to projects that cannot demonstrate they will actually consume the power they reserve.
For enterprise technology leaders, the strategic message is to stop treating grid access as an assumption and start treating it as a scarce, contested resource that must be secured early and defended. The winners in the next phase of AI infrastructure will be the operators who lock in interconnection rights, bring flexible or onsite generation, and accept curtailment terms competitors will not. Texas just made that calculus explicit, and the rest of the country is likely to follow.
