150 Dead Girls: Trump Says Iran Bombed Them and Hegseth Backs Him

This is far worse than anything in the Epstein files.

Yesterday, in speaking with journalists on Air Force One, President Donald John Trump claimed that the Islamic regime in Iran had bombed Shajarah Tayyebeh Girls’ School in the town of Minab near the Strait of Hormuza, resulting in the deaths of over 150 girls and as many 175 people total. (Iranian authorities say 168.)

A reporter asked Trump “did the United States bomb a girls’ elementary school in southern Iran in the first day of the war and kill 175 people?”

Our commander-in-chief responded contradicting reports from The Wall Street Journal and Reuters of likely American culpability. He said that “no, in my opinion, based on what I’ve seen, that was done by Iran.”

The president thinks you are so stupid that you will believe Iran bombed the girls’ school during the US and Israeli attacks.

When the reporter followed up with Defense Secretary Pete Hegseth asking him to confirm the president’s claim, he waffled, stating “we’re certainly investigating” and hedging that “but, the only side that targets civilians is Iran.”

Trump reaffirmed his previous claim, saying that “we figure it was done by Iran, because they’re very inaccurate, as you know, with their munitions. They have no accuracy whatsoever. It was done by Iran.”

---

As if the question of Trump and Hegseth lying about this weren’t enough to turn your stomach, there’s another question going around:

Is the reason for the bombing of the girls school because of the overuse of Anthropic’s Claude chatbot?

Futurist asked the Pentagon if Claude had been used in the attack and they refused to say, prompting AI expert Maria Sukhareva to lay out the facts on Substack Notes:

If the Pentagon refuses to say, we need to look at the evidence ourselves:

1) Unprecedented scale and speed

2) A target database that appears not to have been updated since the school’s separation from the IRGC complex in 2016

3) Confirmed use of AI to generate target packages (summaries explaining why a given target should be struck)

4) The Trump administration significantly loosening requirements for human supervision of the kill chain

5) Reproducible reasoning by Claude that relies on obvious visual markers (flat roof, fence etc.) but lacks the contextual judgement a human military analyst develops over years of experience and applies to subtle, ambiguous cases.

According to The Washington Post, Claude is integrated into the Defense Department’s Maven program and generated 1,000 potential targets which included “precise location coordinates” and also “prioritized those targets according to importance.”

Leading AI thinker Gary Marcus, author of 2024’s Taming Silicon Valley: How We Can Ensure That AI Works for Us, explained at his Substack today that “the thing about including humans in the loop is that if the AI is selecting 80 targets an hour, as has been reported, the humans probably aren’t going to do a good job verifying those targets. Instead, as I warned in a January 2023 essay on driverless cars, ‘[not-quite-perfect] Automation is a double-edged sword, …. the closer [systems] get to perfect, the easier it is for mere mortals to space out.’”

Marcus repeated a message he’s been urging for years, that “overtrust in AI is a huge problem. We have seen it before in law (hallucinated cases that wind up in submitted briefs), journalism, and so on. Now we are (likely) seeing it in war.”

---

Back in November I announced that I’d “fired” OpenAI from our publishing company and would no longer use it myself, explaining how an author pitching me a book based on the program’s so-called “hallucinations” was the last straw:

Since then I’ve spent a lot of time testing the other popular chatbots and LLMs, including Claude.

I’ve downloaded a bunch and organized them into a single folder that I can access from the main homescreen of my iPhone. This way it’s easy for me to take the same question and feed it into multiple apps to see how they differ:

I do not have Grok or DeepSeek as I already regard them as totally worthless. And you don’t see Google Gemini on there because it’s on my home screen itself, below the Drudge Report app and to the left of LinkedIn.

And Gemini is the only one there because it’s the only one that is even quasi-trustworthy because it’s really just a turbo-charged google search that sometimes makes up stuff.

Last year when I was very much still enthralled with the LLMs and on board with the “all in on AI” movement, I didn’t bother with Gemini. ChatGPT was more than sufficient and Gemini just seemed boring.

And so it is! It doesn’t pretend to be some sentient lifeform or the answer to all of our problems. It’s just a search engine with extra bells and whistles. And that’s all I’m needing now. I’m no longer interested in having any sort of “AI companion” or “co-pilot” supposedly trying to help me. And I absolutely will not test any of the “AI agent” programs which can take control of your computer and “accidentally” delete all your files.

And what of the other chatbots in the folder? They’re nothing special or useful. They’re just ChatGPT imitators which put their makers’ personalities over it instead. Do you want a Mountain Dew or a Dr. Pepper? Do you want a Sprite or a 7 Up? Change the flavor all you want, it’s all just carbonated sugar water.

It’s time to get real here:

These apps are just toys.

And so it seemed so surreal to me watching Hegseth try to bully Anthropic’s CEO Dario Amodei into letting the military use Claude to spy on American citizens and deploy no-humans-in-the-loop lethal weapons.

Claude is not reliable enough for me to deploy it in my start-up book publishing company but it’s so essential to our defense that Hegseth must threaten Amodei with bankrupting his business to get full access to it for the most powerful military in human history?

How does that make any sense?

I suppose it makes about as much sense as choosing a sexual predator to become the secretary of defense, a choice affirmed by 50 US senators, necessitating Vice President JD Vance to cast the deciding vote.

These girls’ blood is on your heads too, senators and Mr. Vice President. You put someone totally unqualified into a position of absolute power to kill and this is the result.

---

No one should feel any sort of surprise that men who do not care about the safety of the women and girls in their own lives would not care about the lives of the brown-skinned women and girls on the other side of the world.

Men who cannot tell the truth about themselves will likewise lie about the harm their imperial warmongering inflicts on the whole world.

Trump, Hegseth, and their sycophants can spew their falsehoods all they want. The truth about who is responsible for the deaths of the students and teachers at Shajarah Tayyebeh Girls’ School is more than clear now and will only be further confirmed.

God is watching you, Pete Hegseth. And He knows what you’ve done.

---