Texas Governor Greg Abbott has ordered a comprehensive audit of the state’s data center interconnection queue, effectively pausing grid hookups for hundreds of proposed AI facilities amid grid reliability concerns.
Texas Gov. Greg Abbott has ordered regulators to audit data centers seeking access to the state’s power grid, signaling that even one of the country’s most business-friendly states demands greater scrutiny before massive new power demands come online. According to Quartz, the order pauses the first step of the Electric Reliability Council of Texas’ (ERCOT) revised review system for large power customers.
ERCOT is currently evaluating requests for 474 gigawatts of new demand—more than five times the state’s record peak load—with Governor Abbott noting that approximately 90 percent of those requests originate from data centers. This move aligns with a broader national pushback against large-scale data center development. New York became the first state to halt the construction of new hyperscale facilities statewide, while local officials in areas ranging from Minneapolis to rural Alabama have approved temporary construction slowdowns. These developments are particularly notable in Texas, which has historically been highly enthusiastic about attracting artificial intelligence and technology investment.
By intervening, state officials are expressing concern that poorly prepared projects could clog the grid interconnection pipeline and leave residents vulnerable to associated costs and consequences. ERCOT has also warned, as Quartz noted, that some demand projections may be inflated. In certain instances, companies have secured positions at multiple sites as a strategic hedge, while others appear to have entered the queue before securing financing, land, or a viable electricity strategy. While the audit is unlikely to block major corporations from building, it may filter out weaker or speculative proposals that entered the process prematurely and lack readiness to proceed.
Opposition to data centers has drawn rare bipartisan support. Quartz cited Gallup polling indicating that 70 percent of Americans oppose having a data center built in their community, with nearly half strongly opposed. A separate Data for Progress poll found that 63 percent of voters support pausing data center construction for at least one year. Massive facilities strain local power grids, consume vast quantities of water for cooling, alter local landscapes, and often result in higher utility bills for residents. In many regions, communities view the AI infrastructure buildout less as a local economic benefit and more as an industrial project imposed upon them.
Artificial intelligence and the power grid are inherently intertwined. AI tools can forecast electricity demand, optimize grid operations, and facilitate cleaner energy integration, yet the servers powering these tools require enormous amounts of electricity and water. As AI deployment accelerates, communities are carefully weighing its potential advantages against risks such as resource depletion, security vulnerabilities, potential misuse, and unintended social costs. Political figures are beginning to respond to this growing anxiety. In June, Representative Alexandria Ocasio-Cortez introduced the House companion to a federal moratorium bill, stating that Washington has “a moral obligation to stand with the American people and stop the expansion of these data centers.”
In Texas, the immediate focus is the audit itself. Regulators are working to distinguish genuinely viable projects from those merely occupying queue space without a realistic pathway to construction. This filtering process could enable regulators to concentrate resources on developments capable of moving forward. Meanwhile, other jurisdictions are slowing development through moratoriums or local pauses to thoroughly examine land-use policies, utility impacts, and environmental concerns. In some markets, developers must now post substantial financial deposits before securing a position in the power queue.
These procedural delays inevitably increase project expenses by raising financing costs, jeopardizing land acquisitions, and creating opportunities for lenders or investors to withdraw. Nevertheless, stricter screening may ultimately spare communities the burden of funding projects that were never likely to materialize. Data center developers are now being required to demonstrate that their projects justify the associated trade-offs. Rather than assuming growth is automatically beneficial, regulators and voters are demanding transparent answers regarding power consumption, water usage, local environmental impact, and cost allocation.
The debate over data centers is increasingly centered on who will bear the financial and social costs of the AI boom. Tennessee State Representative Justin Pearson captured this public frustration by describing the situation as “a computer being given more care and concern than a human.”