When Political Theater Meets Geopolitical Reality

Shark Tank investor Kevin O'Leary doesn't mince words when it comes to AI infrastructure, and this week he took aim at Senator Bernie Sanders' suggestion that America should pump the brakes on data center construction. O'Leary's take? That's not just bad policy—it's handing China the keys to the future.

The Canadian businessman posted his response on X Thursday, managing to both compliment and roast Sanders in the same breath. "I love Bernie Sanders. He's the master of the sound bite and the best $5-at-a-time social media fundraiser I've ever seen. He says outrageous things, they go viral, and the donations roll in. That's the playbook," O'Leary wrote.

But admiration for Sanders' fundraising prowess doesn't translate into policy agreement. O'Leary argued that calls to slow technology development or impose moratoriums on data centers are disconnected from how global competition actually works. "That's not protecting workers, that's surrendering leadership," he said. In his view, hitting pause on construction wouldn't safeguard democracy or jobs—it would simply mean someone else builds instead.

The Infrastructure Arms Race Nobody's Talking About

Here's where O'Leary's argument gets interesting. He warned that if the U.S. delays AI infrastructure, that investment doesn't just disappear—it moves to countries willing to build faster. And China, he noted, is more than happy to fill that void.

While American policymakers debate permits and environmental reviews, Beijing is scaling up AI capacity at breakneck speed. O'Leary framed this as more than an economic issue—it's about national security. The country that dominates AI infrastructure could gain advantages in everything from autonomous weapons systems to geopolitical leverage in hotspots like Taiwan.

The numbers tell part of the story. As of November 2025, the U.S. hosts 4,165 data centers—far more than any other nation—while China has 381, according to Statista. But raw numbers don't capture the whole picture. Nvidia Corp. (NVDA) CEO Jensen Huang has pointed out that major AI facilities can take years to complete in the U.S., while China moves much faster on large-scale infrastructure projects. O'Leary has repeatedly hammered this point: American red tape is creating a speed advantage for competitors.

Sanders' Concerns About Power and Profit

To be fair, Sanders isn't worried about the same things O'Leary is. The Vermont independent has argued that billionaire-driven AI expansion threatens jobs, democracy, and public resources. He's called out tech leaders including Elon Musk, Jeff Bezos, Mark Zuckerberg, Peter Thiel, and Bill Gates, saying AI development is being pursued to concentrate wealth and power rather than help working families.

And his concerns about resource consumption aren't unfounded. Experts note that data centers already account for roughly 5% of U.S. electricity consumption, a share expected to climb rapidly as AI adoption accelerates. That's a real cost that communities bear, often without seeing corresponding benefits.

Where Does This Leave Us?

The debate highlights a tension that's become central to technology policy: how do you balance legitimate concerns about inequality and resource use with the reality of global competition? O'Leary's position is essentially that you can't unilaterally disarm in a tech race and expect good outcomes. Sanders' position is that racing ahead without considering who benefits and who pays isn't sustainable either.

What's clear is that the U.S. finds itself in an uncomfortable spot. It has the most data centers but not necessarily the fastest path to building new ones. It has concerns about energy consumption and concentrated wealth but also worries about falling behind China. Threading that needle—building responsibly while building quickly enough to maintain leadership—is the kind of challenge that doesn't lend itself to sound bites or viral tweets.

For what it's worth, Nvidia (NVDA) stock maintains a strong price trend across short, medium, and long-term timeframes, suggesting investors remain bullish on AI infrastructure demand regardless of the political debate.