Summer Associate Matthew J. Wisniewski authored this article with supervision from the Content Editor.

Many believe that the booming data center industry presents a great opportunity for America to be the leader in cutting-edge artificial intelligence innovation; however, potential environmental hazards along with enormous energy demands (and possible increased energy costs) have raised concerns. Numerous stakeholders, including state legislatures and the data centers themselves, are actively working to address these concerns. With Federal regulations recently rolled back, state government officials are now in a position to impose regulations on data centers. But self-regulation may be a more practical, efficient, and long-term solution.

The Supreme Court greatly altered the relevant federal regulatory framework by a 5-4 decision in 2023 in Sackett v. EPA. In Sackett, the Supreme Court narrowed the definition of “Waters of the United States,” thus restricting the power of the EPA to regulate wetlands on which data centers may be built. After the Supreme Court decision, the number of wetlands that were subject to regulation under the Clean Water Act has drastically decreased. As a result, the Trump administration has now decided to leave data center regulation to the states.

At the same time, public opinion on data centers is certainly fractured. On the one hand, data centers objectively provide an opportunity for job growth, millions in tax revenue for states, and a sustained infrastructure for the new age of cloud computing and artificial intelligence. On the other hand, opposition voices cite increased energy costs combined with potential environmental concerns. Due to the overall federal rollback of wetland regulations, citizens who oppose data centers are now shifting their focus to individual states to pass laws and regulations for data centers.

In Pennsylvania, for example, Governor Shapiro recently signed a budget requiring data centers and PJM Interconnection, LLC (PJM) to compile annual reports containing information including “estimated average amount of energy usage per hour during the data center’s peak load.” This provision thus targets increased transparency as well as projections for the energy demand to facilitate resource planning for the anticipated influx of data center loads. A penalty of $10,000 per day will be assessed for data center companies who do not submit the report as directed.

More aggressively, New York is positioning itself to be at the forefront of passing stringent legislation and regulations on data centers by recently becoming the first state to impose a one-year data center moratorium, thus completely halting construction of new data centers with 50 megawatts or greater loads. The threats of increased energy costs and environmental concerns are behind New York’s imposition of this moratorium. The state plans to use this time to draft, review and enact regulations on data centers. As of May 2026, there were 12 gigawatts of very large energy-using loads, including data centers, set to connect to the state’s grid. For context, one gigawatt of electricity can power 750,000 homes.

Conversely, West Virginia is positioning itself as one of the most data center-friendly states in the country. For example, construction has begun on what will be one of the nation’s largest data centers – Nscale’s 8-gigawatt Monarch Compute Campus. This follows West Virginia's passage of HB 2014 in 2025, restricting local authority (particularly zoning authority) to regulate data centers and promoting micro-grid arrangements to support these large load customers, in part to mitigate the potential impact on retail electricity customers. Nscale still needed a stormwater construction permit from West Virginia, but it did not need any state permits for wetlands or water quality.  

With legislative and regulatory oversight of data centers now falling on the states, there is obvious disparity among the states in terms of passing or foregoing regulations.

Some argue that increased regulation may not be necessary because companies that construct data centers have an economic and environmental interest aligned with the public. First, data centers are pledging to make “community-first” infrastructure, promising not to raise energy costs long-term. Microsoft CEO, Satya Nadella, argues that AI strengthens communities in the short and long-term by creating jobs and investing in innovation. Other CEOs are likewise welcoming the discussion on how to best implement their infrastructure while also minimizing environmental harms. Data centers also have a substantial financial incentive to cut their own water and energy consumption.

Second, new technological developments may significantly address environmental concerns. For example, Microsoft has introduced a closed loop for its data centers that helps facilities operate with virtually net-zero water consumption. This technology, similarly employed by other data centers, allows for water to be filled only once, and recirculates the water to cool the data center processes. Data centers have also pledged to take community concerns seriously and to work with stakeholders to address those issues. Ultimately, as noted by those advocating for data centers, it is in the long-term financial and reputational interest of data centers to cut their own energy costs and carbon emissions as much as possible.

Paradoxically, AI actually may provide the solution to climate change, rather than be part of the problem. AI presents an opportunity for innovation that could unleash the full potential of clean energy. Thus, hindering data centers in the name of environmental preservation may end up thwarting critical innovation of clean energy technology that could actually support those environmental goals.

Data centers present a great opportunity for technological development, but no one – not even the data centers themselves – denies their potential environmental impact. Data center-driven AI can solve incredibly complex problems and may play an integral role in the development of clean energy technology. If states stymie data center productivity with too stringent regulations, critical innovation may be hindered or prevented. Conversely, if states deregulate data centers, their high energy costs and water consumption could bottleneck the national energy industry by raising prices through constricted capacity, lowering immediate water supply in critical areas, and increasing overall carbon emissions. This complex dichotomy will continue to be a formidable challenge for states and stakeholders for the foreseeable future.