AI, Data Centers, and the Politics of Control

Originally published on Substack. 

Let’s cut to the chase: AI is not perfect. Neither is government. That should be the starting point.

The real question is not whether AI, data centers, or markets can make mistakes. They can. The question is which system is better at discovering what people want and correcting mistakes.

History says profit and loss beat political permission. Markets punish bad bets. Consumers walk away. Investors lose money. Competitors improve. Government failure usually brings larger budgets, more rules, and another intervention to fix the last one.

That matters because the backlash against AI and data centers is now bipartisan.

Different Parties, Same Control

Consider the strange coalition forming.

New York Gov. Kathy Hochul imposed a moratorium⁠ on new hyperscale data centers. Pennsylvania Gov. Josh Shapiro imposed new restrictions⁠ and removed data centers from fast-track permitting. Texas Gov. Greg Abbott has pushed a statewide review and pause⁠. Texas Attorney General Ken Paxton now wants criminal liability⁠ for data centers when downstream AI chatbots allegedly harm children.

How is that meaningfully different from the interventionist approach of radicals like Texas State Representative and U.S. Senate candidate James Talarico or socialist U.S. Senator Bernie Sanders?

The rhetoric differs. The governing instinct does not. The left distrusts profits and growth. Parts of the right now distrust tech and growth. Both increasingly reach for pauses, moratoriums, mandates, liability schemes, subsidies, zoning limits, and political allocation. That is control.

And Paxton’s proposed liability theory is especially dangerous. Holding a data center criminally liable for downstream chatbot behavior is like holding an electric utility criminally liable for what someone does with an appliance plugged into the grid. The practical result would not be “accountability.” It would be crushing legal exposure that entrenches only the biggest firms.

Small innovators would get wiped out. Trial lawyers would feast. Consumers would pay more. That is not how to let people prosper.

Government Helped Inflate the Boom

There is another uncomfortable truth. Government helped distort this market before politicians turned around and complained about it.

The CHIPS and Science Act⁠ during Biden directed billions toward favored semiconductor investments. States and localities showered projects with tax abatements. President Trump embraced Stargate⁠, a private-sector AI infrastructure venture announced at the White House that could invest up to $500 billion over four years, and said his administration would help accelerate development. (OpenAI⁠)

I welcome private investment. I do not welcome government tilting the field. And yes, the current AI buildout could prove partly bubbly.

We have seen distorted booms before. Housing was badly distorted before 2007. Banking repeatedly expanded under government guarantees and regulatory privilege. The dot-com boom before 2001 mixed real innovation with enormous mispricing.

The lesson is not that technology was fake. The internet clearly was not. Housing and finance did not disappear either. The lesson is that easy money, subsidies, guarantees, tax favors, and political steering can push capital into the wrong places.

If AI investment is overdone, markets should reveal that through losses. Government should not first subsidize the boom and then centrally plan the bust. Two government failures do not make a right.

Profit and Loss Beat Political Permission

This is the classical liberal answer. Protect property rights. Enforce contracts. Make firms pay identifiable infrastructure costs. End special subsidies. Expand energy and water abundance. Then let people experiment.

Some AI firms will fail. Some data centers will be uneconomic. Some applications will be useless. Good. Loss is information.

Government has no equivalent mechanism. Politicians can be spectacularly wrong without losing their own money.

God Gave Us Minds to Create

There is also a deeper reason for optimism. Yesterday brought GPS, search engines, online banking, and smartphones. Today brings AI assistants, smarter logistics, medical tools, robot vacuums, and instant translation. Tomorrow could bring better personal assistants, new ways to communicate, learn, travel, work, heal, and live on God’s earth that we cannot yet imagine.

God gave us brains to discover, create, and improve our lives so we can better serve Him by serving one another. Why would our default be fear?

Technology is not our master. Government should not be either. The proper order is liberty first, innovation through voluntary exchange, and targeted remedies later if actual harms arise.

Bottom Line

I am not “pro-AI.” I am pro-people in markets. Private property and profit-loss are better judges of what people want than taxes, bans, mandates, and political committees.

AI may create risks. So will government efforts to control it. Between the two, I trust free people more than politicians. Liberty and prosperity, not control.

Let markets work. Let people build. Let people innovate. Let people prosper.