Will AI Kill Us All?

A top AI employee just revealed they believe there is a greater than 10% chance that AI “kills us all” by the end of the decade. Needless to say, Trump’s response isn’t reassuring. Here’s the story.

There is perhaps no industry in America that has received a warmer welcome from Donald Trump than artificial intelligence.

Trump has embraced AI with an evangelical fervor he usually reserves for skyscrapers, ballrooms, and anything to do with himself. He has promised to make America the world’s AI superpower (at any cost), stood alongside tech oligarchs to announce hundreds of billions of new dollars to tech and AI companies, and repeatedly insisted that the country needs to build massive data centers – opposition be damned.

That alone would be concerning enough. But this week, the people actually building that technology gave us a rather horrifying reason to ask whether Trump’s embrace of AI is becoming something much more dangerous than his latest billionaire-backed economic crusade against the working class.

Last week, a researcher who spent the last 3 years working at OpenAI and Anthropic, resigned.

In announcing his resignation, he said that AI companies are “gambling with our lives,” despite the fact that the people building AI “earnestly believe that it could kill us all by the end of the decade.”

You heard that right. Experts building the AI models hundreds of millions (if not billions) of people use on a daily basis – ChatGPT, Gemini, Azure, Claude, and more – are convinced they’re quite literally designing humanity’s own demise.

In response, a current Anthropic employee tasked with “alignment science,” agreed saying “we really do earnestly believe AI could kill all humans!” He said Anthropic was the only company that could be trusted with such a technology, and so they were racing to build it to prevent it being used for ill.

Which is pretty much the same logic that led us to develop the atomic bomb, setting off 50 years of nuclear arms races?

As if it couldn’t get any worse, the Anthropic employee also conceded that none of the major tech companies have a plan to solve the problem of aligning artificial intelligence with human interests.

Nor are they even working on a solution.

And now, the people running those companies are asking Washington to do something about it.

Over the weekend, Anthropic CEO Dario Amodei called for the industry to “pace the frontier,” sounding the alarm in saying that AI development is moving faster than our ability to make it safe. Within hours, CEOS at OpenAI, Google, and DeepSeek all publicly endorsed the call for a slower pace of development. They’ve asked for legislation mandating testing and for the government to step in and determine whether sufficiently dangerous AI systems should be blocked.

To put a finer point on that: the CEOs of some of the most powerful companies in the world are telling us that their own technology is advancing too quickly and that they don’t have the capability to regulate themselves without government intervention.

What has Donald Trump’s response been to all of this?

Pedal to the metal. Creative destruction. Gamble with our wealth, our health, and our very lives.

On Monday, Trump rejected the calls for new guardrails, declaring that the only control AI needs is a “STRONG AND SMART (High IQ!) PRESIDENT” and calling efforts to regulate AI and data centers a “SICK conspiracy.” (Trump also came out in favor of flock cameras at the same time, by the way.)

And if AI does rise up?

“We’ll be fine,” Trump says.

Somehow, we’ve landed in the timeline where the people with the strongest reasons to avoid regulation are now the ones begging for it, and the man tasked with passing regulation is not only refusing to lift a finger, but promising to unilaterally block it.

Lately, data centers have emerged as a focal point of the AI development race. Without them, the biggest AI models currently hurtling forward wouldn’t be able to. Or at least not quite as fast. Among Americans from every walk of life or political affiliation, they’ve also drawn near-universal opposition.

In typical Trumpian fashion however, he’s decided to ignore your fears and push on because it’ll make him and his billionaire buddies richer – even if it kills us all in the process.

And if you oppose them?

Trump calls you “backwards” and “poor.”

You definitely wouldn’t oppose them because of valid concerns over the environmental, economic, or social impacts of haphazardly throwing up data centers that consume enormous amounts of electricity and water. Or because of the fact that AI is advancing faster than any technology ever while running entirely unchecked by a Congress full of old octogenarians who have no idea how social media platforms make money (it’s ads), much less any of the intricacies of artificial intelligence.

And of course, you definitely wouldn’t oppose them because these tech companies’ own employees are saying it could kill all of us.

Anyways, according to Trump, what any “rational” person should do is “Let Data Reign.”

Trump isn’t describing AI as one industry among many competing for investment, or for land, or for energy resources. He’s describing it as our economic destiny – or our collective fate.

What Trump really means, though, is we must let data reign over us. If Trump had his way, he’d keep his billions, his houses, his jets, and his golf courses, while we sacrifice, suffer, and lose in the name of further enriching the special interests that have bought the keys to 1600 Pennsylvania Avenue.

In July 2025, Trump signed an executive order directing the federal government to facilitate the “rapid and efficient buildout” of AI data centers and related infrastructure.

Trump’s goal was clear: build more data centers, build them now, and remove any obstacles standing in their way.

More than a year later, data from Erin Brockovich suggests there are over 300 data centers currently under construction, receiving permitting, or pending approval.

In a year.

Erin Brockovich’s Data Center Tracking Website

Though perhaps the most important factor that often gets missed is:

The American people won’t benefit in any meaningful way.

The companies building America’s AI infrastructure are some of the wealthiest on the planet, led by the richest people in the world. The trillions of dollars expected to be generated by AI development won’t be going to regular people trying to get by; it’ll be going to the wealthiest among us, serving only to exacerbate income inequality and erode your buying power in a time where things already cost too damn much.

American communities can’t benefit from wealth they don’t own. But they can absolutely be harmed by it.

This week, we’ve also learned that these same models have been used by the Houthis to autonomously direct missiles to strike U.S. positions in the Middle East. We’ve been warned by those at the very top that, if left unregulated, another AI testing zone breach could bring down the entire internet. We’ve seen universal agreement (if you ignore McDonald’s biggest customer) that AI humanoid robots could turn on humans and wipe us out entirely.

To call that merely scary would be the understatement of the century.

Today, Trump stands on an island of one. It isn’t hyperbole to say that our historically unpopular age-addled, weight-rattled 80-year-old president is the only person standing against AI regulation – ‘till death do us part, apparently.

Whether you’re a Democrat or Republican, in favor of or opposed to AI, or frankly couldn’t care less about any of it, we can all agree that the federal government shouldn’t be flagrantly ignoring the will of everyday Americans in favor of the same under-regulated, under-taxed oligarchs when even they are now pleading for regulation.

Even when the experts and CEOs tell us AI may kill us all, what’s Trump’s response?

Let Data Reign.

With this president, the only game he’s interested in playing is Monopoly – except we’re already 10 turns and hundreds of billions of dollars behind.