The stakes are high in the Pentagon’s battle against ‘woke AI’

It seems like a soap opera, or a pitch for a Hallmark movie. But it’s all too real, and the future of our military could hang on how it all plays out.

The U.S. war against Iran is one of the first in the realm of modern warfare where artificial intelligence is playing a key role in all aspects of the conflict, from planning to propaganda. It will not be the last.

How this goes, and how effectively the Pentagon is able to use these new tools, will set the tone for the next 100 years of warfare, like it or not. So now is not the time for a conflict with an artificial intelligence provider — wars are hard enough when taken just one at a time.

The week before the start of hostilities, the Trump administration made news by issuing an ultimatum to Anthropic, an AI company with Pentagon contracts, demanding that it either allow the military unfettered use of its Claude AI or else risk having them seize it through the Defense Production Act — a law that gives the federal government wide latitude in the name of national security.

Anthropic refused, and at just the wrong moment, when the Defense Department was turning to Claude to prepare its attack on Iran. Ultimately, Secretary Pete Hegseth decided not only that his department would not use Claude, but also that it is a supply-chain risk that all defense contractors must avoid going forward.

The real question isn’t Claude’s usefulness — one has to assume the Pentagon knows what it is doing in the making of war, and what gives it an advantage. The real question is why a company like Anthropic, evidently suffering moral angst over how its product would be used, would have ever done business with the Pentagon in the first place. 

Anthropic is concerned that our military could use its tech to “undermine, rather than defend, democratic values.” If so, it should never have submitted a bid. But the more profound question is that of who makes decisions about what America’s military must do. Will it be our elected leaders, as the idea of democracy would seem to require, or will it be unelected entrepreneurs in Silicon Valley with eccentric personal ideas about morality?

The New York Post looked into this in November and discovered Anthropic was tied to what it called the “cult-like effective altruism movement.” Effective altruism, the Post noted, “urges devotees to funnel their cash to causes they believe will most benefit humanity in the future. Dubbed ‘long-termism’ by followers, those causes include climate change, preventing pandemics — and a near-obsession with heavily regulating artificial intelligence.”

But as we saw with one of the most famous effective altruists — Sam Bankman-Fried — this odd brand of morality was just an excuse to create Political Action Committees and donate to Democrats who would help him financially benefit.

Does that sound compatible with the Trump administration’s ideal of America First? or does it sound like “woke” in sheep’s clothing?

Nathan Leamer, the CEO of Fixed Gear Strategies, describes effective altruism adherents as believing “that only a handful of people can control access — that AI, if not in their hands, will be leveraged to destroy the world. … Effective altruism is a governing philosophy that is entirely built on godless progressive ideas” — and, I would add, on very undemocratic ideas, at that.

This drama tellingly involves a cameo (at least so far, who knows how deep the rabbit hole actually goes?) by the late billionaire and sex-criminal Jeffrey Epstein. Anthropic CEO Dario Amodei used to work at OpenAI, where he was apparently working on a projected called “The Deep Thinking Project.” This had been brought to Epstein’s attention by a literary agent named John Brockman, whom The New Republic dubbed “Jeffrey Epstein’s intellectual enabler.”

Just how involved Epstein was in anything related to Anthropic’s development, or in the idea of effective altruism, is unknown. But his appearance in the background is a reminder of just how small the AI universe really is. It should also prompt you to chuckle whenever you hear of tech leaders lecturing others on morality as a driving force in anything business-related.

The singular priority of the military must always be to conduct of war effectively on behalf of the U.S. and its interests. If artificial intelligence can aid in that, then great — but the contractors cannot be allowed to dictate the terms on which their product contributes. If they don’t want to aid in the mission, that is their right, but in that case they should not take the government’s money in the first place. Once they do take it, the flow of command orders can only go in one direction.

Anthropic took the money and developed the product. It doesn’t get to retain control based on its developers’ views on what constitutes “altruism.”

AI can be as “woke” as it wants to be in the private sector, but in the military, the mission has to come first.

Derek Hunter is host of The Derek Hunter Show, weekdays from 3 p.m. to 6 p.m. Eastern on WMAL Radio in Washington. He is a former staffer for the late Sen. Conrad Burns (R-Mont.).

Copyright 2026 Nexstar Media Inc. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.

More Opinions - National Security News

The return-to-the-office trend backfires

Live results: Georgia special election to replace Marjorie Taylor Greene

Swalwell says staffer flew to Colombia to return hearing aid to deported ...

Speaker Johnson floats using reconciliation to address alleged fraud in blue ...

Thune quashes Trump push to reform filibuster for SAVE Act

Democratic senator says US ‘seem to be’ on path toward deploying American ...

Republicans eye second GOP-only megabill, but it faces uphill climb

Odds of a ‘Super El Nino’ growing, model shows. What that could mean

Erika Kirk appointed to Air Force Academy board

Rand Paul: Midterms will be ‘disastrous’ for Republicans

Trump heads to Kentucky in an aggressive effort to topple a foe: GOP ...

Projectiles hit 3 ships in Strait of Hormuz

Thompson fends off challenger to win Mississippi primary

2 states approved permanent standard time. Others are hoping to do the same

Ro Khanna faces primary challenge, Silicon Valley backlash over wealth tax

Joe Rogan: Trump supporters ‘feel betrayed’ by ‘insane’ Iran war

Iran war flummoxes some Trump media allies

Fuller, Harris nab spots in Georgia House runoff for Greene’s seat

The Hill Podcasts – Morning Report

© The Hill