Joe Biden handed Silicon Valley’s most ideologically driven AI company a sweetheart deal.

Now the man cleaning up that mess is telling the whole story.

And what the Trump Pentagon discovered will leave you shaking your head at what the last administration let slide.

More Questions About AI

Emil Michael, the Under Secretary of War for Research and Engineering and Department of War Chief Technology Officer, sat down with the Ruthless Podcast to walk through the chain of events that ended Anthropic’s relationship with the Pentagon. The interview pulls back the curtain on a contract so badly negotiated that the man currently running the Pentagon’s AI portfolio couldn’t believe what he was reading when he finally got around to it.

“I got the [artificial intelligence (AI)] portfolio,” Michael told the hosts. “Six months into the job, I read the [Anthropic] contract. ‘You can’t use AI for battlefield management.’ I was like, ‘What does that mean?’ ‘You can’t use AI to defend yourself from a missile attack.’ What?'”

Read that again. The United States military signed a contract with an AI company that explicitly prohibited using that AI to defend American troops from missile attacks. That’s not a minor oversight buried in the fine print. That’s a fundamental betrayal of what a defense contract is supposed to accomplish.

Michael, who previously served as chief business officer at Uber, didn’t mince words about who was responsible. “[This contract was from] the Biden administration,” he quipped. “Maybe someone didn’t read it.”

And that’s the charitable interpretation. The less charitable one is that somebody read it just fine and didn’t care, because the Biden White House had already, according to Michael, used its computer regulations to effectively crown Anthropic as the “chosen AI company” for the federal government. A Silicon Valley company steeped in effective altruist ideology got preferential treatment from a Democrat administration, and the terms of that preference apparently included letting the company’s moral philosophy override the operational needs of the United States military.

When Anthropic Told the Pentagon It Had Its Own Constitution

Michael didn’t just read the contract and walk away. He called Anthropic in and tried to fix it.

“So I called [Anthropic] in, and I was like, OK, you guys, you know, got to change this,” Michael told the hosts. “We’ll use it lawfully, for all lawful purposes, but it’s artificial general intelligence. You don’t get to tell me where I could fly a plane to. We decide when we buy a plane from Boeing or Lockheed.”

Anthropic’s response to that reasonable request says everything about how this company thinks about its relationship with the American government. “And they’re like, ‘We think these are our moral values. We have our own constitution.'”

Their own constitution. The company that the Biden White House anointed as the Pentagon’s preferred AI vendor told the United States Department of War that it answers to its own moral document rather than to the needs of the country it was contracted to serve. Podcast co-host Comfortably Smug noted that this posture is spreading across Silicon Valley. “These are E.A. people who are the effective altruists who are just like a menace essentially,” Smug said of the growing trend.

He’s not wrong. Effective altruism is the philosophical movement that produced Sam Bankman-Fried, the FTX fraudster who convinced himself that stealing customer funds was justified by utilitarian calculations about future good. The same ideological ecosystem that rationalizes fraud as philanthropy is now baking its moral preferences into AI systems that the United States military is expected to depend on in combat. The idea that a software company’s internal ethics document should constrain how American soldiers defend themselves is not a principled stand. It’s a power grab dressed up in academic language.

The Pentagon wasn’t asking Anthropic to do anything unlawful. Michael’s position was explicit: use the AI for all lawful purposes. Anthropic refused. And the refusal came wrapped in the language of moral superiority from a company whose CEO, Dario Amodei, has publicly warned that AI could eliminate half of all entry-level white-collar jobs and drive unemployment to 10 to 20 percent within one to five years. The same man sounding alarms about what his technology will do to American workers was simultaneously insisting that the American military couldn’t use that technology to protect American lives.

What Happens When Big Tech Decides It Outranks the Pentagon

This spring, the Department of War froze its relationship with Anthropic and signed new contracts with SpaceX, OpenAI, Google, NVIDIA, Reflection, Microsoft, and Amazon Web Services. The Pentagon’s move is currently being litigated in federal court, with Anthropic publicly arguing against the Department of War’s authorization policy to use the technology for “all lawful purposes.”

But the legal fight is almost beside the point. The more important question is how a company ended up with a Pentagon contract that restricted what the Pentagon could do with its own purchased technology in the first place. And the answer to that question runs straight through the Biden administration’s habit of treating ideologically aligned tech companies as partners in governance rather than vendors subject to the same terms as any other contractor.

Think about what the effective altruist crowd actually believes. They believe they are uniquely positioned to calculate the long-term consequences of powerful technology, and that their calculations should override the preferences of elected governments, military commanders, and the people actually responsible for national security. That’s not humility. That’s the same arrogance that convinced Sam Bankman-Fried he was smart enough to steal and still come out ahead on the moral ledger.

And it carries real national security risk. Anthropic’s Claude AI has been woven into Pentagon supply chains and operational systems. A company whose leadership is ideologically committed to restricting military use of AI, and which has demonstrated it will litigate rather than comply when the government asks for full lawful access, is not a reliable defense contractor. It’s a liability.

The Biden White House built that liability into the foundation of the Pentagon’s AI infrastructure. The Trump administration is now paying the legal and operational costs of unwinding it.

There’s a broader pattern here worth naming. The same Big Tech companies that built massive data centers raising electricity bills for working families, the same companies that throttled conservative speech on COVID policy and the 2020 election, are now asserting the right to tell the United States military what it can and cannot do with the tools it purchases. The effective altruist movement that animates Anthropic’s leadership didn’t materialize from nowhere. It grew up inside the same Silicon Valley culture that treats its own moral preferences as superior to democratic accountability.

Anthropic’s CEO has warned, in his own words, that the technology his company sells could cause “unusually painful” disruption to the American workforce. He has said AI will have effects that are “much broader and occur much faster” than prior technological shifts. And yet his company’s position is that the United States military should be restricted in how it uses that same technology to defend the country.

Emil Michael’s interview on the Ruthless Podcast is the clearest public accounting yet of how a Biden-era contracting failure turned an ideologically driven AI company into a de facto veto holder over Pentagon operations. The Trump administration moved to fix it. Anthropic responded with a lawsuit. The court battle will play out, but the underlying dynamic is already settled: a private company decided its own constitution outranked the security needs of the United States, and the Biden White House let them write that assumption into a federal contract.

That’s what the cleanup looks like. And it’s only getting started.

Source: Fox News, “Ruthless exclusive: Pentagon CTO reveals reasoning for Anthropic removal”