> We saw in a movie how motorcycles can jump over bridges. We used
AI to learn how to do this. We gave it information, like what
motorcycles we use and the distance we need to jump and so on and it
gave us steps on what we have to do. We practiced a lot and kept
asking questions. We dug holes and filled them with broken glass and
fire to practice. 18 of us died in the process. Eight of us managed to do
it. The next time we attacked, we could jump.
Now listen, I'm not saying we need to give these guys more AI, but it clearly isn't yielding bad outcomes for us here.
"You're absolutely correct! For it to be a good practice ground you need to fill the trenches with broken glass and light the whole thing on fire"
Not gonna lie, I'd rather attack with 26 fighters that haven't survived lots of jump attempts than 8 who are much more confident in their motorbike stunt riding but presumably still aren't bulletproof.
But maybe they could ask Claude how to train themselves to resist bullets as well?
You type in the question or use your voice and it [AI] gives you a detailed answer, like ‘How can I build a bomb?’ and then it tells you how. It is like a human robot! We used it a lot.
I’m pretty skeptical reading this bit. I’ve seen uncensored or jailbroken LLM replies to these kind of questions, they are never actionable, don’t say anything Wikipedia doesn’t, and are hard to provoke if you’re not using an uncensored model.
I have no doubt terrorists are aided by LLMs in a general sense, but am skeptical of any claim that they are providing some material embargoed knowledge that isn’t available elsewhere, in a way that either improves efficiency or effectiveness of their activities, and would want to see real evidence, not an interview snippet.
I read stuff like this and think I must be an idiot because I'm so bad at circumventing the AI safety for fairly benign queries. And here you have folks making bombs...?
Making explosives is generally fully legal in the US, so IDK if there would even be safeguards for US hosted AI, since there's no real legal issue with doing it. Basically no federal regulations for non-commercial production so long as it isn't stored or moved anywhere, you can literally buy tannerite off the shelf in a sporting good store, "synthesize" it by mixing it and then blow up a huge bomb legally, no license required. YouTube is plastered with people inside the USA making TNT and other materials and then blowing them up.
I can believe some would, but some probably wouldn't care. My local ranch store certainly will happily sell anyone tannerite without even a background check or any sort of scrutiny, you can cash and carry it. Walmart won't but the point being as long as it's legal there will be a "ranch store" that carries it.
I mean even YouTube allows bomb making videos and they won't even usually allow videos of people making guns. It's just not very regulated in the US enough to make most companies care.
There's lots of knowledge out there about stuff like this. Milennia of humans tinkering with things that go boom. Surfacing it more easily has value (in a manner of speaking; as the @dril tweet goes, "you do not, under any circumstances, 'gotta hand it to them'").
Yeah Dugan Ashley recently went to jail because a terrorist found his RDX synthesis video on YouTube (where it and other videos had been for years) and used it (incorrectly) in NOLA. The CIA literally released to public domain how to make the blasting caps and pack the primary explosives, the US and foreign governments released the full synthesis of primary and secondary explosives from the patent office...
I can't imagine what assistance AI would have been for them, other than maybe translating it to whatever terrorist language they were using.
I agree with other commenters that the claims made in the report are strange.
> We used to rely on our traditional methods. We sent 200 fighters because we had a lot of strength, but then 60 got killed. With the help of AI, we learned that it sometimes makes sense to only send 20. We learned more about well-coordinated attacks and deployment of smaller units.
The other quotes and use cases could make sense in terms of using AI jailbreaks to find information more easily, but this one is absolutely ridiculous. Did the clueless researcher just get trolled?
Or the researcher read what they wanted to into it. It would be interesting to ask them what they did before to learn things, how much they read, etc. If they were illiterate and uneducated, and got voice AI telling them stuff that would be common sense for anyone with a high school education, I can see how it might make them more effective at whatever they do. But I wouldn’t really blame AI in the way that’s implied.
Human brains were shaped over thousands of years of adaptation for warfare. Just look at how creative and advanced the tactics of other guerrilla forces (like the Taliban and Viet Cong) got, despite their very limited resources. None of that needs a high school education.
Their brains are the same, what differs is the environment.
Humanity's superpower is the ability to copy and mirror each other very effectively.
It does not require advanced awareness or intelligence to do it. Most people copy others subconsciously!
Majority of people won't contribute anything technologically, but they sure as hell can copy.
> We saw in a movie how motorcycles can jump over bridges. We used AI to learn how to do this. We gave it information, like what motorcycles we use and the distance we need to jump and so on and it gave us steps on what we have to do. We practiced a lot and kept asking questions. We dug holes and filled them with broken glass and fire to practice. 18 of us died in the process. Eight of us managed to do it. The next time we attacked, we could jump.
Now listen, I'm not saying we need to give these guys more AI, but it clearly isn't yielding bad outcomes for us here.
"You're absolutely correct! For it to be a good practice ground you need to fill the trenches with broken glass and light the whole thing on fire"
Nobody tell them that there are models that work much better!
Not gonna lie, I'd rather attack with 26 fighters that haven't survived lots of jump attempts than 8 who are much more confident in their motorbike stunt riding but presumably still aren't bulletproof.
But maybe they could ask Claude how to train themselves to resist bullets as well?
I have no doubt terrorists are aided by LLMs in a general sense, but am skeptical of any claim that they are providing some material embargoed knowledge that isn’t available elsewhere, in a way that either improves efficiency or effectiveness of their activities, and would want to see real evidence, not an interview snippet.
How do they bypass the AI safety measures?
I read stuff like this and think I must be an idiot because I'm so bad at circumventing the AI safety for fairly benign queries. And here you have folks making bombs...?
Making explosives is generally fully legal in the US, so IDK if there would even be safeguards for US hosted AI, since there's no real legal issue with doing it. Basically no federal regulations for non-commercial production so long as it isn't stored or moved anywhere, you can literally buy tannerite off the shelf in a sporting good store, "synthesize" it by mixing it and then blow up a huge bomb legally, no license required. YouTube is plastered with people inside the USA making TNT and other materials and then blowing them up.
Have you tried chatbots? They invoke AI safety for lots of (very) legal things. The whole point is not to allow people to make bombs.
Legality has nothing to do with it.
I can believe some would, but some probably wouldn't care. My local ranch store certainly will happily sell anyone tannerite without even a background check or any sort of scrutiny, you can cash and carry it. Walmart won't but the point being as long as it's legal there will be a "ranch store" that carries it.
I mean even YouTube allows bomb making videos and they won't even usually allow videos of people making guns. It's just not very regulated in the US enough to make most companies care.
There's lots of knowledge out there about stuff like this. Milennia of humans tinkering with things that go boom. Surfacing it more easily has value (in a manner of speaking; as the @dril tweet goes, "you do not, under any circumstances, 'gotta hand it to them'").
Yeah Dugan Ashley recently went to jail because a terrorist found his RDX synthesis video on YouTube (where it and other videos had been for years) and used it (incorrectly) in NOLA. The CIA literally released to public domain how to make the blasting caps and pack the primary explosives, the US and foreign governments released the full synthesis of primary and secondary explosives from the patent office...
I can't imagine what assistance AI would have been for them, other than maybe translating it to whatever terrorist language they were using.
I agree with other commenters that the claims made in the report are strange.
> We used to rely on our traditional methods. We sent 200 fighters because we had a lot of strength, but then 60 got killed. With the help of AI, we learned that it sometimes makes sense to only send 20. We learned more about well-coordinated attacks and deployment of smaller units.
The other quotes and use cases could make sense in terms of using AI jailbreaks to find information more easily, but this one is absolutely ridiculous. Did the clueless researcher just get trolled?
Or the researcher read what they wanted to into it. It would be interesting to ask them what they did before to learn things, how much they read, etc. If they were illiterate and uneducated, and got voice AI telling them stuff that would be common sense for anyone with a high school education, I can see how it might make them more effective at whatever they do. But I wouldn’t really blame AI in the way that’s implied.
Human brains were shaped over thousands of years of adaptation for warfare. Just look at how creative and advanced the tactics of other guerrilla forces (like the Taliban and Viet Cong) got, despite their very limited resources. None of that needs a high school education.
You send 200 fighters and 60 get killed. In the same situation, if you send 20, what do you expect would happen? (I mean, you won't lose 60...)
Sort of fascinating how Bronze Age cultures can co-opt our technology.
Can't let their underage harem girls dress normally, but they can follow the instructions to make a dirty bomb.
Maybe we need to do a better job isolating them, or at least not making it so easy to follow.
Their brains are the same, what differs is the environment.
Humanity's superpower is the ability to copy and mirror each other very effectively. It does not require advanced awareness or intelligence to do it. Most people copy others subconsciously!
Majority of people won't contribute anything technologically, but they sure as hell can copy.
I noticed the nytimes just published an article about this.
How Terrorist Groups Are Using A.I. to Gain an Edge in Battle https://www.nytimes.com/2026/07/10/us/politics/ai-terrorism-...
If you opened the OP link, you missed the section "Featured in The New York Times".
We need to ban open source AI for regular citizens to prevent terrorists from using them.
Mandatory ID verification at software level for local LLMs is clearly the solution here. /s
I would be more interested about terrorists organization like Al-Shabaab that at least control many towns.
Does Boko Haram and ISWAP even control a single town or they just control a few villages in Lake Chad and in the Sambisa forest?
Also reading the report they seem quite clueless.