Why New York said no to AI Data Centres?

For fifty years, the story of American industry has been the story of where electricity is cheap. Aluminum went to the Northwest. Steel went to the coalfields. Data centers went wherever the tax breaks were fattest. Yesterday, one state decided that party's over. And the ripple could reach further than anyone's willing to say out loud.

But before we get to that, let's take a quick look at the markets and what matters…

AI is fueling the Fourth Industrial Revolution – and these 7 stocks are front and center. 

One of them makes $40K accelerator chips with a full-stack platform that all but guarantees wide adoption. 

Another leads warehouse automation, with a $23B backlog – including all 47 distribution centers of a top U.S. retailer – plus a JV to lease robots to mid-market operators. 

From core infrastructure to automation leaders, these companies and other leaders are all in 7 AI Stocks to Invest in Today.

Free today, grab it before the paywall locks.

And with that out of the way, let's get to today's big story: the moratorium heard around Silicon Valley.

For most of the twentieth century, the story of American industry was the story of where the electricity was.

Aluminum smelters followed hydro dams to the Pacific Northwest. Steel followed coal into western Pennsylvania. Later, semiconductor fabs and rare-earth processors began chasing cheap kilowatt-hours around the map like herds.

The pattern was so consistent that industrial economists gave it a name. Load-following. Factories moved to the power. Not the other way around.

And then, for about twenty years, the whole thing went silent.

Between 2005 and 2020, total US electricity consumption barely budged. Utilities got so used to flat demand that some of them forgot how to plan for anything else. Coal plants were retiring. Nuclear was contracting. No one was building new baseload because no one thought they'd need it.

Then AI arrived.

Somewhere around 2023, the demand curve went vertical again. Not from factories. Not from population growth. From boxes full of GPUs.

The Electric Power Research Institute now projects that data centers could consume 9 to 17% of all US electricity by 2030. That's roughly double where they sit today. A country the size of Spain, plugged in overnight.

And yesterday, New York decided it had seen enough.

Governor Kathy Hochul signed an executive order pausing state environmental permits for any new hyperscale data center for up to one year. The threshold is 50 megawatts or more. It's the first statewide moratorium of its kind in the country.

Officials framed it as an environmental review. Water. Cooling. Air quality. All the usual paperwork.

But you can find the real reason two clicks deeper on the governor's own press release, where they quietly promise the pause will "ensure New Yorkers are not paying for transmission and infrastructure build outs."

That is the tell.

Trump has signed 220 Executive Orders in one year…more than almost every U.S. president in history.

Now, on July 24th… He’s preparing to sign what sources say will be his final one.

A White House leak suggests this won’t just erase Biden’s legacy…

It will trigger a $2 trillion initiative to radically reshape America forever.

While making fortunes for those who are prepared for what’s coming.

The details are shocking. But you can’t miss this.

See what Ian King uncovered.

Under the rules that govern most US electricity markets, when a hyperscaler wants to plug a new gigawatt-scale campus into the grid, the utility has to build new substations, transformers, and high-voltage lines to serve it.

Those costs are, in theory, paid by the customer that requested them.

In practice, under existing tariff structures, they get socialized across the entire ratepayer base, unless explicit rules say otherwise. Every household. Every small business. Everyone.

The scale is not small.

In the PJM grid region, which spans thirteen states, data centers accounted for a $9.3 billion capacity-price increase in the 2025-26 auction alone, according to the market monitor's own analysis. Average residential bills are expected to rise $18 a month in western Maryland. $16 a month in Ohio.

Virginia is the extreme case. Areas with heavy data center concentration have seen electricity prices jump 267% over five years. Dominion Energy filed its first base-rate increase since 1992 earlier this year.

Ratepayers noticed.

The physical world has finally sent AI an invoice.

Well, New York already has more than 12 gigawatts of large energy users, mostly data centers, sitting in its grid interconnection queue. That's enough to power roughly nine million homes.

The state already has the eighth-most expensive residential electricity in the country. It doesn't have Texas's spare wind or Ohio's spare gas. And it has a governor in a tight re-election fight against an opponent who wants to keep the projects coming.

The math wrote itself.

You see, this isn't really about environmental review. It's about who bears the cost of the AI buildout. 

Big Tech has collectively committed over $320 billion in data center capex in 2025 alone, roughly double what the entire US electric utility industry spent on generation and transmission that year. The physics has become the constraint. And the physics gets paid for by whoever the tariff rules say pays.

For a decade, the AI story was told entirely in software. Model weights. Parameters. Training runs. It was a story that lived in the cloud, in the abstract, in the code.

Yesterday's moratorium is what happens when that story meets a substation.

Every gigawatt of AI compute now requires a physical world response. A transformer that takes two to four years to manufacture. A transmission line that takes a decade to permit. A community that gets to vote, one way or another, on whether it wants to host the thing at all.

The Trump administration has been trying to speed this up. 

The Department of Energy is accelerating permits. FERC is reforming interconnection queues. The White House has run "ratepayer protection pledge" ceremonies asking hyperscalers to voluntarily cover their own costs.

None of that gets around the local politics.

Because the deepest constraint on the AI industry is no longer chips. It's not talent. It's not even capital. It's the willingness of a chap to pay eighteen extra dollars a month so a data center he'll never see can train a model he'll never use.

And that willingness, it turns out, has limits.

Three Mile Island had its 1979. Nuclear power spent thirty years in the wilderness after that. Not because the technology stopped working. Because the social license to build ran out.

AI is now getting its first real taste of the same medicine.

Whether New York's moratorium turns out to be a hiccup or a turning point depends on one thing: whether hyperscalers can figure out how to pay their own way before another dozen states copy the executive order sitting on Governor Hochul's desk.

Because in the end, no infrastructure survives its own bill.

If you have money in the markets, Public Law 63-43 could have a huge impact on your wealth in the second half of 2026.

Three words buried deep in Section 10 of this 112-year-old law may give President Trump the power to make a critical move.

A former advisor to the CIA, Pentagon, and White House says meetings are already happening behind closed doors.

See the move he believes could trigger a historic wealth supercycle.

Today's reply prompt: Do you think other states follow New York within six months? Hit reply and tell us why or why not.

Until then, sip slowly!

The Market Sip Desk