The Need for AI Vs. the Need for Protecting Land, Water, and Bank Accounts

One state finds itself different when it comes to all the rest. That state is New York. Communities across the country are having passionate debates – for and against – about locating data centers in their jurisdictions. But New York initiated a temporary ban for the entire state.

More than a dozen states have discussed the idea of temporary bans, according to the National Conference of State Legislatures (NCSL), which has been tracking legislation.

But none progressed as far as New York.

Data center moratorium

New York Governor Kathy Hochul signed an executive order, which establishes a one-year moratorium on “hyperscale” data centers. The executive order would cover data center projects that require 50 megawatts or more in power, while state leaders design a new regulatory framework.

This first-of-its-kind statewide moratorium comes at a time that opposition seems to be growing as some people question how much data centers could further raise electricity costs, as well as the additional strain they could put on water and other resources.

“New York has always been at the forefront of innovation and change but we’ve also always guaranteed that New Yorkers benefit. As data center development threatens to hike up utility bills, deplete our natural resources, and create uncertainty for New Yorkers, it’s my responsibility to take action and lead,” Governor Hochul said in a statement. 

She continued, “New York will lead the way in creating the strongest standards in the nation for data center development, ensuring that when companies succeed because of New York, New Yorkers succeed too.”

Supporting data centers

Critics contend that blocking data center projects only stunts the country’s ability to meet the demands driven by massive expansion of artificial intelligence and could force the U.S. to fall behind global competitors like China.

Hochul’s actions brought some bipartisan criticism. “China wins,” U.S. Senator John Fetterman (D – PA) posted on X/Twitter.

New York Assemblyman Scott Gray (R-Watertown), ranking Minority member of the Assembly Standing Committee on Energy, said that it makes sense for people to have questions about the natural resources that data centers often require. But he opposed Hochul’s executive order.

“A statewide moratorium is the wrong answer to the right questions. It freezes investment, takes decisions away from the communities that should be making them and duplicates or ignores work the governor’s own administration already has underway,” Gray said in a statement. “The responsible path is to set clear, firm standards and let willing communities decide for themselves.”

Data center future

Artificial intelligence’s demand for support has developed a new level of data center projects, according to an article in Data Centre Magazine (note: as you might have guessed from the spelling of “centre”, this publication is produced by a company that is headquartered in the United Kingdom).

“Conventional data centres draw a steady, predictable baseline of electricity,” the article stated. “AI infrastructure behaves differently, with foundation model training pulling vast amounts of energy over multi-week spikes before dropping to highly variable levels during live deployment. That volatility is what utilities are finding hardest to plan around.”

AI could require 25-60% of total data center power demand within three to five years, according to a report from Capgemini, a Paris-based firm that specializes in technology consulting.

AI data center projects could also demand substantially more power. Data Centre Magazine cited information from the International Energy Agency that a traditional data center consumes a similar amount of energy as 100,000 homes. But an AI data center may require 20 times that amount.

Where will the projects get the power that they need?

Local and state leaders will now have to decide how to balance the need for more data centers as AI expands, while considering the requirements of land (including farms and ranches), water, and infrastructure.

Even AI may not be able to come up with a quick solution for that.