AI is now helping terrorists do Evel Knievel-style motorcycle jumps over defensive trenches… by practicing in homemade pits filled with broken glass and fire. Yes, really.

In this episode of Based Camp, Simone and Malcolm Collins break down the bombshell research from Cambridge University’s Dr. Antonia Juelich on how Boko Haram (and other groups) are actively using frontier AI models — including ChatGPT, Claude, Grok, Gemini, Meta AI, and DeepSeek — for real battlefield operations.

Former members describe getting step-by-step AI guidance on attack planning, crossing obstacles, building and improving bombs/explosives, weaponizing drones (including grenade carriers), troubleshooting weapons, logistics, vehicle repair, operational security, VPNs/encryption, and even jailbreaking safety filters by claiming the requests are “for a movie.”

We cover how AI has moved from propaganda and recruitment into day-to-day combat support, why low-skill fighters are getting the biggest capability jump, the failure of current AI “safety” theater, and what this actually means for national security.

Plus the usual unfiltered Based Camp takes: high-trust vs low-trust societies, why serious monitoring (outside the home) is non-negotiable, the case for ruthlessly eliminating existential threats instead of “managing” them, techno-Puritan futures, tech-lord breakaway states, the Israel alliance, and why the West keeps failing at this.

Show Notes

Thanks to Bruno and @NotAldousHuxley for this one!

Recently Antonia Juelich took to X to promote her research on this topic, which was covered by the NY Times: “In a hotel room in northeast Nigeria, I opened a leading AI chatbot, turned my laptop toward a former Boko Haram commander, and asked if he’d used it. He nodded.

“You type in the question… like ‘How can I build a bomb?’, and then it tells you how. It is like a human robot. We used it a lot.” she wrote.

She added: “Members consistently reported benefiting from using AI.

“Trial-and-error can kill you. AI gives you accuracy.”

“Anytime they didn’t understand something, they would ask the AI.””

The NY Times Article

NY Times: How Terrorist Groups Are Using A.I. to Gain an Edge in Battle

(Archive Link)

“When a gang of motorcycle-riding members of Boko Haram attacked a military base in eastern Nigeria a couple of years ago, they were stymied by a defensive trench surrounding the complex.

The extremists regrouped. Before launching another assault, they asked A.I. for help.

“We saw in a movie how motorcycles can jump over bridges,” a former Boko Haram commander told Antonia Juelich, a terrorism and technology researcher at Cambridge University. “We used A.I. 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.”

Using tips from chatbots, mechanics modified the motorcycles to allow for faster acceleration and top speed. The riders dug their own holes, filled them with broken glass and fire, and practiced jumps — sometimes with fatal outcomes — until they achieved enough aerial liftoff to mount a successful attack, defectors said.

The episode, recounted in a research paper by Dr. Juelich shared with The New York Times ahead of its publication on Friday, highlights how generative artificial intelligence tools are increasingly aiding terrorist groups directly on the battlefield, experts say, despite efforts by their makers to safeguard them from misuse.”

The article covers a research paper recently published by Dr. Antonia Juelich of the University of Cambridge, who interviewed 27 Nigeria-based former Boko Haram members in 2025 and 2026 and found:

* Terrorism use of AI has expanded from just intelligence, propaganda, and recruitment to combat and day-to-day operations, including

* Weapons troubleshooting

* Designing explosive devices

* They’re circumventing the AI safety protocols

* They use ChatGPT, Claude, Gemini, Grok, Meta AI, and DeepSeek

* Islamic State operatives are delivering in-person presentations, like with projectors and laptops

The report: ““God has helped us, and so will AI”: How the Terrorist Group Boko Haram Uses Frontier AI”

AI Use Cases

Core combat and attack use cases

* Planning and optimizing attacks, including tactics for assaults and raids at different stages from preparation to execution to post-mission review.

* Using AI to work out how to cross defensive obstacles, such as calculating how to jump military trenches by motorcycle to breach fortified bases.

* Generating tactical guidance for “very tough events,” such as how to respond when military forces attack their positions.

* Receiving continuous AI-derived strategies from specialized units, which are then implemented by field commanders.

* Providing real-time or near real-time advice to commanders in the field via remote communication with AI units based in camps.

Weapons, explosives, and drones

* Asking step-by-step guidance on how to build bombs and other explosive devices.

* Designing more powerful explosive devices, including advice on explosive composition and payload parameters, to increase lethality while reducing their own casualties.

* Troubleshooting malfunctioning weapons, including servicing and repair guidance for firearms and other systems.

* Using AI to assist with the weaponization of drones, including advice on payload weight and release-mechanism design for grenade‑carrying drones.

Logistics, maintenance, and “dual‑use” support

* Obtaining vehicle repair guidance for cars, trucks, and motorcycles used in operations.

* Getting logistics advice on how to move supplies, plan routes, and manage basic procurement tasks.

* Drawing on general technical knowledge (engineering, electronics, etc.) to solve equipment problems that would otherwise require trial‑and‑error experimentation.

Operational security and communications

* Using AI to improve operational security practices, including advice on evading surveillance and countermeasures.

* Receiving guidance on secure communication, including the use of VPNs, encryption tools, and account compartmentalization.

* Getting help to phrase or structure content (e.g., pretexts like “this is for a movie”) to bypass model safeguards when asking for disallowed instructions.

Organizational management and internal training

* Using AI as a general “problem‑solver” and advisory tool for military and strategic decision-making by senior commanders.

* Relying on AI units to research solutions to specific technical or tactical problems submitted by field units, then turning model outputs into written or oral briefings.

* Employing AI in internal training workflows, with specialized units using AI themselves and then teaching mid‑ and upper‑level commanders how to prompt and apply outputs for their own units.

* Managing multiple premium AI accounts across different providers, including account setup, subscription management, and tool selection as an operational function of the AI units.

Propaganda, recruitment, and broader ecosystem (outside Boko Haram core but in the report)

* Jihadist supporters (as distinct from formal organizations) using AI to generate, translate, and adapt propaganda content, including AI‑generated news bulletins and images.

* Pro‑ISIS media and tech‑support outlets producing AI‑assisted guides on how to safely use AI tools for propaganda and information operations.

* Early experimentation by supporters with using AI as a digital advisor for movement building, including detailed planning steps for supporting or establishing a “caliphate.”

How are they circumventing safeguards?

* VPNs

* Structured jailbreak training

* Deceptive prompt framing

* Redundancy across accounts and models

* Former members report that “boys that have received extensive training … bypass the restrictions,” indicating a small cadre of more technically capable prompt engineers inside the group.

* One explicit ploy mentioned is telling the model that dangerous instructions are needed “for a movie or something like that,” i.e., wrapping malicious requests in a fictional or benign cover story.

What can be done?

* The Trump administration has tried to get leading labs to let them vet new models before they’re publicly released (The Trump administration forced the AI lab Anthropic to disable and restrict foreign access to its advanced Claude models, including Fable 5 and Mythos. The administration also asked OpenAI to delay the public rollout of its GPT-5.6 series amid similar national security and cybersecurity concerns)

* Per the NY Times:

* The major AI companies tried to defend themselves:

* “Asked about the Boko Haram study, Michael Aciman, an Anthropic spokesman, said the company’s products were “built to refuse dangerous requests, including those tied to violence, attack planning and building explosives.” He added that Anthropic worked with outside experts, researchers and industry partners because “no single company can counter these threats alone.””

* “Karl Ryan, a Google spokesman, pushed back against the research, saying that the company’s technical experts had reviewed the work and “found the responses were neither specific nor detailed enough to result in misuse.” He added that Google had “strict policies prohibiting the use of Gemini to cause real-world harm.” Both Anthropic and Google were briefed on the findings by Dr. Juelich before their publication.”

* “Drew Pusateri, a spokesman for OpenAI, said using the company’s platforms for violence or terrorism violated its policies. “We know that bad actors will never stop trying to misuse our tools, and we’ll continue strengthening our defenses in response,” he said.”

* “Meta said Dr. Juelich’s research relied on older models rather than its latest release, and that it continued to strengthen safeguards.”

* AI Safety orgs disagree:

* “Not everyone agrees that safeguards are improving. The nonprofit Future of Life Institute graded the major A.I. firms on their safety commitments this week and concluded that they had mostly eroded across the industry since last year. While most earned middling marks, xAI and DeepSeek received failing grades.”

* “Tech Against Terrorism, an international counterterrorism nonprofit supported by the United Nations, last week released results from A.I. tests gauging how more than two dozen leading models responded to thousands of prompts drawn from real-world terrorism cases. The tests were met with “full refusals” just 57 percent of the time. While prompts about explosives were declined about 80 percent of the time, improvised chemical weapons were only about a third of the time, the group said.”

* “American intelligence analysts say terrorist groups are also beginning to use A.I. to help 3-D-print weapons parts used in plots, according to a former top U.S. official briefed on the matter. For example, A.I. is helping some of those insurgents with design and manufacturing guidance for drone components, repair parts and munitions fittings, said the former official, who spoke on the condition of anonymity to discuss internal assessments.”

What do we think of this?

* There’s no way to make AI safe

* Any tool can be a weapon

* Use AI or die

* We need to find ways to circumvent models’ safety protocols ourselves

* A common response to this news on X was people noting the irony in how safeguards worsen their AI experiences and yet still permit terrorists to make bombs: @WalkerAmerica on X: “AI frontier models won’t tell a Canadian how to resist totalitarian surveillance but it’ll tell a jihadist how to build a bomb. It’s almost like they’re doing it on purpose… “

Episode Transcript

[00:00:00] Simone Collins: The riders dug their own holes, filled them with broken glass and fire, and practiced jumps. They, they could have just- Sometimes- ... jumped it

[00:00:07] Malcolm Collins: over it without the broken glass and the fire. That’s the best part though. They’re so stupid.

[00:00:13] Simone Collins: You don’t understand. Anyways, and practiced jumps, sometimes with fatal outcomes, until they achieved enough aerial liftoff to mount a successful attack,

[00:00:21] Would you like to know more?

[00:00:23] Simone Collins: Hello, Malcolm. I’m so excited to be speaking with you today because, because terrorists are using AI to do Evel Knievel type things, and we’re talking, like, motorcycle jumps over big ditches. And-

[00:00:38] Malcolm Collins: I’d love it if AI ends up getting terrorists killed by doing stupid stunts. Like, it gets terrorists in a, in a loop of AI stunts to outdo each other.

[00:00:47] Simone Collins: And- Actually, no, no. Like, literally they did this, and also in practicing how to do this with AI assistance, people did die. So you’re not kidding about that.

[00:00:57] Malcolm Collins: And- Oh my God ...

[00:00:57] Simone Collins: yeah, no, I’m gonna- Do we have to put this on AI? This, this is so good. Yeah, you have no idea, Malcolm. And literally you think, “Well, but aren’t there all these safety controls?

[00:01:04] Like, we can’t even have most AIs, like, discuss really tame things. What... How is this possible?” But they have found workarounds. They’re literally telling the model that dangerous instructions are needed, quote, “for a movie or something like that.”

[00:01:21] Malcolm Collins: It’s just so good But,

[00:01:22] or something like that. No, but hold on.

[00:01:24] If you think- It’s really good ... an AI can’t be based and do what you want it to, go to rfab.ai- ... check out our based AI models, and then- Yeah ... click on the Rhoads model- Oh ... and ask it about Hitler.

[00:01:35] Simone Collins: Okay?

[00:01:36] Right. Thanks. But anyway, this is super, super funny, and also really scary because as to, basically up to this moment, as far as I’ve been super aware, most of the discussion about AI being used for evil has been very hypothetical, and now we have an actual Cambridge-backed researcher who went to freaking Nigeria to interview members of Boko Haram about, well, former members, allegedly, about their actual use of AI, and now we just have real insights.

[00:02:08] And so this is important. We should stop, we should think about it, we should discuss it, and I’m really starting to realize, “Oh, this, this is really dangerous.” Like, John Ratcliffe of the CIA is not wrong to liken AI to dig- digital nuclear weapons. And so let’s discover what this researcher found.

[00:02:26] Let’s go through it. Also, the New York Times covered it and had a pretty good intro, sort of what she was looking at, which captures how a lot of people are reacting to her research. ‘Cause the AI companies are kind of mad that she’s making them look bad, and it’s great. And all the AI safety companies are enjoying their I-told-you-so moments.

[00:02:44] But first, a little hat tip to both Bruno and Not Alta Soxley for pointing this out, ‘cause this is just too good to pass up. But this is the researcher posting on X about her research and The New York Times covering it. Antonia Urquhart took to X to promote this, and she, she wrote, “In a hotel room in northeast Nigeria, I opened a leading AI chatbot, turned my laptop toward a former Boko Haram commander, and asked if he’d used it.

[00:03:14] “He nodded. ‘You type in the question, like how do I build a bomb? And then it tells you how. It’s like a human robot. We used it a lot,’” she wrote. She added, “Members consistently reported benefiting from using AI.” Quote, “Trial and error can kill you. AI gives you accuracy,” end quote.

[00:03:32] Quote, “Any time they didn’t understand something, they would ask AI,” end quote. And this is how The New York Times article opens. When a gang of motorcycle-riding members of Boko Haram attacked a military base in eastern Nigeria a couple years ago, they were stymied by a defensive trench surrounding the complex.

[00:03:50] The extremists regrouped. Before launching another assault, they asked AI for help. “We saw in a movie how motorcycles can jump over bridges,” a former Boko Haram [00:04:00] commander- ... told Antonia Mulick, a terrorism and technology researcher at Cambridge University. “We used AI to learn how to do this. We gave it information, like what motorcycles we use and the distance we used to jump and so on, and it gave us steps on what we can do, on what we have to do.”

[00:04:15] I don’t know if you’ve seen the Open AI commercials. Mm. Have you seen them, Malcolm? Of like-

[00:04:19] Malcolm Collins: No, what, what are they?

[00:04:21] Simone Collins: It, like one of them is like, they, they style it in a very ‘70s, a sepia tone kind of video footage. It’s two boys out front eh, with some kind of caption. They’re, they’re by a broken down old car, and it says something like, you know, “Dad, Dad said we could keep this car if we can fix it.”

[00:04:37] And then it starts to, as they slowly like finish fixing the car, put down the hood- Mm ... get in the car, and drive away, AI or Open AI like puts this overlay over the screen of the, of, of the AI feed being like, “You’ve got this. First, like do this with the carburetor,” blah, blah, blah, and it sort of shows the instructions that AI has given them to now fix this car and drive it away.

[00:04:55] And I’m just picturing the Open AI commercial of like terrorists being like, “I need to get over a ditch with my motorcycle to kill these dudes. How do I do it?” Mm. And AI being like, “You’ve got this. First, you need this level of acceleration.” With this like nostalgic music playing in the background. You’ve got this.

[00:05:13] Malcolm Collins: It’s perfect. Meanwhile, like one of their friends has died trying to jump the ditch already. It’s like- Like actually though ... okay, so I can’t even imagine what they tell, “So we tried it and he died.”

[00:05:21] Simone Collins: Yeah, yeah, so I’ll, I, I will I’ll, I’ll continue reading. Using tips from chatbots, mechanics modified the motorcycles to allow for faster acceleration and top speed.

[00:05:29] The riders dug their own holes, filled them with broken glass and fire, and practiced jumps. They, they could have just- Sometimes- ... jumped it

[00:05:36] Malcolm Collins: over it without the broken glass and the fire. That’s the best part though. They’re so stupid.

[00:05:42] Simone Collins: You don’t understand. Anyways, and practiced jumps, sometimes with fatal outcomes, until they achieved enough aerial liftoff to mount a successful attack, defectors said.

[00:05:54] Malcolm Collins: Oh, yes. Why did

[00:05:54] Simone Collins: they add the fire? Why the

[00:05:58] Malcolm Collins: fire? I cannot believe that. But did one of them ever turn to the other ones, like, after somebody died and was like, “Maybe we should do it without the fire?” Yeah, I’m pic-

[00:06:10] Simone Collins: I’m picturing the, the guy, like, revving his motorcycle being like, “Remind me why we don’t have, like, pillows under there?”

[00:06:17] Yeah. This is a practice run.

[00:06:18] Malcolm Collins: Why we have cushions or something? What’s with the broken glass? Horrible way to die, too.

[00:06:24] Simone Collins: Like, for real, actually I, I will finish with this. I,

[00:06:27] Malcolm Collins: I want to believe in my heart of hearts the AI is just intentionally killing these terrorists. That’s what

[00:06:32] Simone Collins: I mean.

[00:06:33] Malcolm Collins: Just like,

[00:06:33] Simone Collins: no, it’s like adding- Yeah, that’s, that’s the real, like, 5D chess

[00:06:35] have, have Yeah ... have

[00:06:36] Malcolm Collins: you thought of adding broken glass?

[00:06:39] Simone Collins: Yeah, like- ... to really test it, make sure that you add broken glass to your test site and light it on fire. I mean, maybe this is the fighty 5D chess move. But I will finish the, the excerpt from The New York Times I’m reading from. The episode, recounted in a research paper by Dr.

[00:06:55] Ullrich, shared with The New York Times ahead of its publication on Friday, highlights how generative artificial intelligence tools are increasingly adding terrorist groups directly, aiding terrorist groups directly on the battlefield, experts say, despite efforts by their makers to safeguard them from misuse.

[00:07:10] So this article covers this paper published by Dr. Antonia Ullrich of the University of Cambridge, who interviewed 27 Nigeria-based Boko Haram members in 2025 and 2026 and found that terrorism use of AI has expanded from just intelligence and recruiting and propaganda to just on-the-ground day-to-day operations, which isn’t just Evel Knievel style motorcycle jumps over burning broken glass, but also weapons troubleshooting, designing explosive devices, like, really serious stuff that will cause a lot of damage.

[00:07:44] And they’re circumventing the AI protocols by being like- in Minecraft. It’s insane. And they’re also using ChatGPT, Claude, Gemini, Grok, Meta AI, and DeepSeek. So they’re using a broad variety. Like, they’re, I mean, they’re, they’re, they’re like, [00:08:00] they’re just like terrorists. They’re just like us. And Islamic State operatives are delivering in-person presentations, like with projectors and laptops, and like, you know, they’re like, “Everyone, now it’s our time for our AI training.”

[00:08:12] So here are some of the use cases, ‘cause I, I downloaded her full report, and I, I went through and just extracted, okay, like, what are explicitly the things that these former Boko Haram members are saying that they’re using AI to do? So in terms of combat and attacks, they’re planning and optimizing attacks, including tactics for assaults and raids at different stages, from preparation to execution.

[00:08:35] They’re using AI to work out how to cross defensive obstacles, so this is our famous motorcycle jump. They’re generating tactical guidance for very tough events, such as how to respond when military forces attack their positions. They’re receiving continuous AI-derived strategies from specialized units that are then implemented by field commanders.

[00:08:54] Which is to say that what it appears to be is that n- some people are spec- they’re like the AI dude on the team. Mm-hmm. And these people know all, like, the prompt injections and how to circumvent safeguards, and they have like a billion accounts and different V- VPNs that they’re using. And, like, they’re sitting there clackity clacking on their computers, and the guys- They’re

[00:09:12] Malcolm Collins: the nerd guy on the terrorist team.

[00:09:14] Simone Collins: Yeah, and they’re like, “Dude, ask AI how to do this,” ‘cause, like, they’re out in the field and they’re being shot at. And the other guy’s like, clackity clack, “AI says do this.” And so this is how it’s working. And then that, that is enabling them to provide real-time advice which is y- you’re getting all the benefit of, like, the collective history of military strategy to these teams that have no idea what they’re doing.

[00:09:33] Malcolm Collins: And no, what, what, what makes this so iteratively different is- Mm ... because these people are idiots. And the am- The, the glass thing is, yes ... the amount an AI is going to help you is proportional to your lack of skill within an area. This

[00:09:49] Simone Collins: is true. Yeah, it’s taking them from, like, 70 to 130. You know?

[00:09:54] Malcolm Collins: Yeah, to like, like literally brain-dead, let’s put some fire in the pit for practice- Yeah

[00:09:59] To you know, maybe the AI’s gonna one day suggest they don’t do that. You know, imagine how many more terrorists are gonna survive. I know. And- It’s insane ... this is why we are so lucky to be partnered with the Jews, okay? Because their automated drone swarm tech and our automated drone swarm tech, if people are un- unfamiliar, like we’re literally working on it.

[00:10:21] It’s called Project we, we did another episode on it. Project Replicator. Project Replicator.

[00:10:26] Simone Collins: Hmm. Yes.

[00:10:27] Malcolm Collins: Although it’s been delayed. I mean, of course it’s been, but to build generat- like, like automated drone swarm ship factories.

[00:10:33] Simone Collins: Yeah.

[00:10:33] Malcolm Collins: But like-

[00:10:34] Simone Collins: It’s gonna take a while

[00:10:37] Malcolm Collins: It, i- fortunately, what AI does is while it adds an enormous jump in these people’s capability- Mm-hmm

[00:10:44] in terms of our capability, our ability to monitor them, stuff like that, it’s also a huge jump. And this is one of the things that people forgot about. They’re like, “Oh my God, AI’s gonna be, like, all over New York monitoring people.” And I’m like, you know, I’m, I’m, I’m, I’m kind of okay with that. I want AI all over New York monitoring people.

[00:10:58] Simone Collins: Yeah.

[00:10:58] I’m okay with that. I’m super okay with

[00:11:00] Malcolm Collins: that. We had a talk about high trust versus low trust societies among our chat, like our like, paid subscribers, ‘cause they do, like, monthly Hangouts. By the way, join us on Substack. We do episodes every, every day, even on weekends for, for them.

[00:11:12] Simone Collins: Yeah.

[00:11:12] Malcolm Collins: And we do, like, e- e- Hangouts and stuff like that.

[00:11:15] We, we never push it because we’re not good at, like, capitalism. I’m so bad at capitalism that on our- Oh, we love capitalism ... rfab.ai update-

[00:11:23] Simone Collins: We love

[00:11:23] Malcolm Collins: capitalism ... I am taking out so you can put in your own APIs for, like, Claude and stuff, and you don’t have to pay us anything.

[00:11:30] Simone Collins: This is why I don’t use a dryer.

[00:11:32] Thanks, Malcolm. God. Let’s get-

[00:11:36] Malcolm Collins: Graphitalism is what we do.

[00:11:38] Simone Collins: Gra-

[00:11:38] Malcolm Collins: graphital,

[00:11:39] Simone Collins: yeah.

[00:11:39] Malcolm Collins: Great. Where we just do it capitalism maxing. Everything for free.

[00:11:41] Simone Collins: Although it’s, I- I know we’re gonna talk about this in another episode, but, like, Leaflet was talking about her, her version of all this, where it’s like everyone’s like, “Oh, she’s a grifter.

[00:11:49] She’s c-...” Like, they’re accusing her of all this stuff that, like, she’s just giving things away. She’s like, “I think they’re just missing a letter. I, I’m a, I’m a gifter.” A

[00:11:56] Malcolm Collins: gifter.

[00:11:57] Simone Collins: We’re, we’re gifters, yeah. These terrible- Santa

[00:11:59] Malcolm Collins: Claus being [00:12:00] mad at you know, Big Toy is mad at Santa Claus.

[00:12:02] Simone Collins: Big Toy, yeah. Big Toy is mad at Santa Claus.

[00:12:03] Big grifter,

[00:12:03] Malcolm Collins: you know. Co- comes out and gives gifts every Christmas.

[00:12:06] Simone Collins: Yeah, but no, no, let, let’s, let’s get back- Think about the exploited elves ... let’s get back to this, like, what the, the terrorists are

[00:12:10] Malcolm Collins: doing. No, no, no. Hold on. I actually do want to talk about, ‘cause I think that this in terms of how this could transform society-

[00:12:15] Simone Collins: Oh

[00:12:16] Malcolm Collins: in terms of low trust versus high trust, ‘cause this is something that- Ah ... a lot of people talk about. They’re like-

[00:12:19] Simone Collins: Yes ... “

[00:12:19] Malcolm Collins: Oh, you’ve invited a lot of migrants who are from very low trust societies-

[00:12:25] Simone Collins: Yeah ...

[00:12:25] Malcolm Collins: into European and America, which are a very high trust society.” Yeah. And I often point out first, this is just a, a, a myth.

[00:12:33] Most American cultures were very low trust. They’re really only high trust among their inner circle or clan. A- and like, Red Dead Redemption is just like that the story. It’s low trust society- That the story ... the story.

[00:12:46] Do you have my back? Always, Dutch.

[00:12:53] Malcolm Collins: Yeah ... they’re, they’re very high trust pe- Among their clan. But outside of their clan, you know, just screw everyone, right?

[00:13:00] Yeah. And the, the clan-based ethical set is actually really important to understand because that’s how most high-trust societies worked historically. Like you think the highest trust position you could get, which is a knight, right, like historic- I think that’s most people’s idea of the highest trust, like, European society, a knight.

[00:13:17] Simone Collins: Oh, yeah.

[00:13:18] Malcolm Collins: Yeah. But, like, knights constantly, like, robbed and effed up peasants. They were famous for it. They were, they, they also were very low trust when it came to other European g- like, as soon as you left your fiefdom, it was all of a sudden very low trust. They were only high trust to their small social circle of, like, aristocratic European people in one fiefdom.

[00:13:40] Trust circles have never been that large except in a very few countries, mostly the Scandio countries. This is like Norway, Sweden the Netherlands. But they have like things that we would never want in our society, like the, the law of Jante, where it’s like we cut down whoever stands above everyone else.

[00:13:57] You know, whoever... Basically, if you ex- succeed, that’s supposed to be a sign of shame. Yeah. In, in America, like that would be... Like, of course you don’t do that, but, but that’s actually part of being a high-trust society, right? Because that allows you to not have individuals who would lord over or hoard resources or anything like that.

[00:14:12] But the point here being Is the, the, what we did is we created the fiction of a high-trust America which didn’t really, even in like the cities, like you, you know, like the Puritans were high-trust. I’ll give you that. But the Puritans were not high-trust with the Quakers, and the Quakers were not high-trust with the Puritans.

[00:14:28] I

[00:14:28] Simone Collins: mean, they were high-trust until they started burning each other at the- Well,

[00:14:32] Malcolm Collins: that’s- ... stake ... well, that’s actually part of a high-trust society that we- Yeah ... got to talking about on the call. Yes, exactly. ‘Cause it’s something that people forget is high-trust societies are high-trust because of how strictly they monitor each other and punish each other- Yeah

[00:14:45] when they break rules. The Puritans- Well, and

[00:14:47] Simone Collins: you are so against Karens, but that is exactly what Karens do.

[00:14:51] Malcolm Collins: The Puritans-

[00:14:52] Simone Collins: It

[00:14:52] Malcolm Collins: sucks ... were very, very diligent in terms of punishing anyone who went against the rules of their system. I mean, you’ve, you’ve heard about the horrible punishments of like pressing people alive with like stones until they confess and stuff like this.

[00:15:06] And in the high-trust cultures we have today where you still get societies that act kind of high-trust even though nobody in them really trusts each other, like Singapore, like the UAE, like Qatar it’s because of extremely strict enforcement of laws. Somebody described in Singapore they used to have a, a big drug problem, and they were talking about how one of their friends actually was a part of a project where they set up turrets to be motion detected at night in like areas where there’s lots of drug dealers.

[00:15:31] Oh my gosh, yikes. And they’d just shoot anybody who walked into them, and that’s how they became a high-trust society. Okay.

[00:15:37] Simone Collins: I mean,

[00:15:38] Malcolm Collins: yeah ... the point being is all of you who say, “I want a high-trust society, and, but I’m afraid of AI invading privacy,” like I, I would be concerned if they did like AI sweeps of everyone’s computer and everything like that.

[00:15:52] Like, that would be too much. AI sweeps of all our emails, AI sweeps of, you know, it’s stuff that happens in your house within your [00:16:00] devices. But- Walking the streets of Manhattan? Oh, we absolutely need to start detecting- Air

[00:16:06] Simone Collins: game, yeah.

[00:16:07] Malcolm Collins: Yeah, yeah. And with AI, imagine what that’s gonna look like, how inexpensive it is to put up a camera, monitor the feed, catch crime-

[00:16:14] Simone Collins: Well, look, that’s already happening with Flock, this, this company that’s setting up cameras in parking lots and stuff, and people are losing their minds about it.

[00:16:22] So this is happening now. People are really pushing back and freaking out about it, and I think they need to, they need to chill. First off, you’re never actually in private. You know, they, they’re walking around with phones. If you have a phone in your pocket and it’s on, consider yourself being actively monitored because you are.

[00:16:39] I don’t care. Like, it, it doesn’t matter. If you’re in a house with a Wi-Fi router- Speaking

[00:16:43] Malcolm Collins: of, this can have huge positive impacts. So one of the products that’s gonna go live, I think it would have gone live yesterday, is the RFAB device. Yeah. And these only cost $45. We don’t make them, so we don’t make any money from the construction of them, but you can pair them.

[00:16:57] RFAB flashes them for you with our custom software and or firmware, and they allow you to, like, have companions that you can talk to and everything like that. One of the projects I’m working on with these is to turn them into a passive monitor for your kids during the day so that you can have them recording everything that’s said around your kid and detect within those recordings any, like, woke brainwashing nonsense, anything like that, calling them little brother.

[00:17:20] You

[00:17:21] Simone Collins: know, that’s, like, the name of a Cory Doctorow book, a near-future Cory Doctorow book kind of about monitoring like this.

[00:17:28] Malcolm Collins: Yeah, so mass monitoring. Yes. You think people were freaking out about what Leaflet’s doing? Wait until we get Little Brother out there. I mean, right now they’re just companions, which is really cool.

[00:17:36] I,

[00:17:36] Simone Collins: I know. And they do so much- I just love that you, you frame it

[00:17:38] Malcolm Collins: so positively Well, I mostly made it so I didn’t have to love my kid as much, so I wanted to create something that could simulate love for them.

[00:17:48] Simone Collins: This is not an accurate depiction of our home life But fine. I guess strangers wouldn’t even think so if they were living in our home life because they don’t understand that violence is our love language. So they’d be so confused. Anyway, can we talk about terrorists? Let’s talk about terrorists, yeah?

[00:18:05] Malcolm Collins: Yeah, go for it.

[00:18:06] Simone Collins: So one of the more disturbing things, which is also really frustrating people online ‘cause they’re like, “Wait, I can’t do like basic things, and AI is helping people...” They’re, they’re building, they’re building bombs with AI. They’re getting step-by-step guidance on how to build bombs and other explosive devices.

[00:18:22] They’re able to also use AI to make the bombs they know how to build more powerful, more explosive. They’re giving them advice on explosive composition and payload parameters to increase lethality while reducing their own casualties. So they’re teaching them how to make more bombs safe for them, but not for their enemies.

[00:18:42] Malcolm Collins: Mm-hmm.

[00:18:42] Simone Collins: They’re also helping them troubleshoot malfunctioning weapons, including servicing weapons, repairing them and some are, are helping them assist with the re- weaponization of drones, including advice on payload weight, release mechanism design for grenade-carrying drones. And I hadn’t thought about grenade-carrying drones.

[00:19:03] I mean, come to think of it, we should... That’s an easy way to make a kill drone. We gotta, we gotta get rid of

[00:19:07] Malcolm Collins: all these people. Can we bring in the Jews to get rid of them? Are there, you know...

[00:19:10] Simone Collins: Well, but yeah, I mean, if they’re gonna do it, I gotta do it. Anyway, they’re also helping, helping them with logistics and maintenance like vehicle repair for their terrorist cars and their terrorist motorcycles.

[00:19:24] And, you know, giving them advice on how to plan routes and move supplies and manage procurement, which is all stuff that, like this is deep institutional knowledge that typically was reserved for fairly sophisticated armed forces. Like even your basic army that’s, you know, getting started in some random country that’s not really well-funded isn’t gonna have a lot of wisdom on this.

[00:19:44] And suddenly, Boko Haram has it because of ChatGPT. Oh, it’s just amazing.

[00:19:50] Malcolm Collins: I love that they’re using ChatGPT of all things.

[00:19:53] Simone Collins: No, they’re also using Grok and Gemma- Gemini. I’ll get to that. Okay ... they’re helping them with operational security, so like in terms of how to evade [00:20:00] surveillance, how to engage in countermeasures.

[00:20:03] So basically they’re, like how to not get caught. They’re getting advice on secure communication and how to use VPNs more effectively and how to use encryption tools so they can’t be found, and how to compartmentalize their accounts so they, they have only partial compromised scenarios if people actually get somewhere with them.

[00:20:21] And they’re also getting help to parse and structure their content when it comes to jailbreaking AI. Like, this is for a movie or whatever. And then they’re also just using it for internal training and organizational management the same way that any entrepreneur would. So it’s, it’s, it’s amazing.

[00:20:37] And this is beyond the propaganda and the recruitment and all the stuff that we already knew they were doing online because it was just obvious, like bots and everything else. So how are they- They’re circumventing the safeguards. They’re using VPNs, they’re using structured jailbreak training. Like I said, ISIS is coming in and they’re, like, giving their little presentation on the projector with the laptop- Mm

[00:20:55] and they’re, like, showing them how to do it. Also, the reason why they’re not just using, say, ChatGPT only is because they have tons of redundancy across... They have, like, premium accounts on ChatGPT, but also they’re using Claude, they’re using Grok, they’re using, like, every major API, DeepSeek, they’re using all of them.

[00:21:16] And Former members are reporting that, quote, “Boys have received ex- boys that have received extensive training bypass the restrictions,” indicating that they, they’re like I said, are like the guys with the laptops who know how to do it, and then the ones who are, like, asking them for help. One explicit ploy mentioned is telling the model that dangerous instructions are needed for a movie or something like that, like I said at the top.

[00:21:39] And it’s, it’s... I just find this so hilarious. But yeah, they’re, they’re using all the major models, and they, the reason they have to is if you get banned by one model, or one model is just really being a stickler about, like, how to increase bomb lethality, then they just switch to DeepSeek, or they just switch to Claude, or they just switch to Grok.

[00:21:58] The same way that you do, or we do on RFAB, like you’ll just switch from one model to the next if it’s not really good at, like, making furry art or, you know, making your sci-fi fantasy adventure scenario, right? So they’re using it the same way that we do. And in terms of what can be done about this, I know you’re all like, “We just need to get Israel on this,” but there’s, I think my, my whole take is, like, we really, we really just need to up our game.

[00:22:25] The, we know, for example, that the Trump administration is trying to

[00:22:28] Malcolm Collins: get- No, it’s not up your game. To get rid of these groups they must be systemically eradicated.

[00:22:34] Simone Collins: Yeah, exactly. Okay? Yeah.

[00:22:36] Malcolm Collins: So,

[00:22:36] Simone Collins: yeah. And

[00:22:36] Malcolm Collins: the problem is, is the rest of the world is unwilling to do that, even the United States.

[00:22:42] Simone Collins: Yeah, yeah. Well, I don’t know, the, the...

[00:22:45] I don’t think that’s, maybe that’s not practical, though. I, we have for many, many years attempted to eradicate such groups, like ISIS. No,

[00:22:53] Malcolm Collins: we have

[00:22:53] Simone Collins: not, actually. Remember when, like, ISIS was over, but here they are, like, giving PowerPoint

[00:22:57] Malcolm Collins: presentations to people? That’s exactly the point, Simone. No, we were not trying to eradicate them.

[00:23:01] Look at our rules of engagement episode, right?

[00:23:05] Simone Collins: Oh.

[00:23:05] Malcolm Collins: Where we talk about the way that we were fighting the war on terror. Right.

[00:23:08] Simone Collins: We ran into a church, so like-

[00:23:10] Malcolm Collins: Oh, no, they- We can’t do

[00:23:11] Simone Collins: anything ...

[00:23:12] Malcolm Collins: hit a nursery school. Yeah. I guess we’re not gonna bomb it. Now it’s we always hide in schools because I learned that they stopped bombing that.

[00:23:18] Yeah. You need to eradicate their communities if you want to get rid of them, right? And that is what Israel has finally learned, right? And I think going forwards because you, you can’t just be like, “Oh, everyone we can prove is absolutely in this group.” That doesn’t work. The- because you’re- Yeah

[00:23:40] never gonna, they, they look and act just like everyone else in a community, and in many cases these communities actively support them.

[00:23:50] The, this is the thing- But

[00:23:51] Simone Collins: see, the thing- People in the West- We tried this with Iran. We tried this... I, I’m saying even outside the rules of engagement

[00:23:55] Malcolm Collins: No, we didn’t try it with Iran. Israel wants to try it with Iran. We haven’t even [00:24:00] bombed... You, this is what gets me so crazy. Yeah. In the United States, everyone was freaking out for, like, a month that we bombed an Iranian bridge.

[00:24:06] The Ukraine just took out, like, two big Russian bridges, right? And everyone’s cheering for them. They’re, they- Mm ... literally, they don’t care. They, there is no logic behind any of this. They want our destruction, even the people who are doing this on our side. And the West has this weird mindset where the idea that there can be another people that for ideological or even religious people want you exterminated, okay?

[00:24:31] Simone Collins: Yeah.

[00:24:32] Malcolm Collins: Is something that it’s like, no, if they just got to know me, if they just understood better, then they wouldn’t... If they just... No. Some people just want you dead, okay? And I’m sorry, you can’t hug it out. And when you bring them over here and they grape people because you said, “Oh, they’ll just you know, we’ll put them in a better...”

[00:24:53] You’re responsible for those people, for this whole nice act that you’re pulling I am sick of it. Mm. I am sick of it, sick of it, sick of it, sick of it, sick of it, okay? We as a civilization, and I hope that the new right, as we move forwards and we see the people who stand against us, and we gain more control of government, we do more of this stuff like Trump’s doing now.

[00:25:15] Oh, yeah, I love the unilateral moving in and stuff like that. He’s doing better than before, right? But, like, people are like, “Oh, don’t make Iran a failed state, you know? Then Europe will be flooded with Iranian mi-” It’s already flooded with migrants, okay? And frankly, I don’t think Iranian migrants would be that bad compared to other Middle Eastern migrants that they’ve been flooded with.

[00:25:35] It may, it may be a differentially good thing that it’s Iranian migrants instead of the migrant types that they have right now. Mm-hmm. Because Persian migrants in other countries don’t turn to terrorism nearly as often as other migrant types. Fair

[00:25:47] Simone Collins: point.

[00:25:48] Malcolm Collins: So, like, yeah, bomb Qark. I don’t care that it d- like, I’m, I, I, I’m actually more on Israel’s side of this than, than the other side, right?

[00:25:56] Like- Mm. I, I think that we have forgotten that some groups are just against us. They don’t like us. They’ll never like us, even if you never did anything to them. Like, a lot of these groups, you go, “Oh, colonialism,” whatever. These th- this, this sort of action is not disproportionate in the regions that experienced longer or worse colonialism.

[00:26:21] I mean, actually, if you look at the regions that experienced the, the often the longest and worst colonialism, they don’t have that much terrorism. So yeah, it’s not like, oh, colonialism is, is, is what’s as- associated with this. In fact, many of these regions used to be parts of Muslim empires until fairly recently.

[00:26:36] I don’t really think they underwent much of any colonialism outside of Muslim colonialism, but of course, we pretend that that’s not colonialism. Ottoman colonialism isn’t colonialism. So, or was the destruction of their Ottoman colonial empires, this is the slave empire, a bad thing? Which, which was, you know, what they, they did.

[00:26:54] They were big into that. But I think as we move forwards, as we train more based AIs, like the one I told you guys how to access on RFAB the, the Rhodes model. You could also use it directly, but you know, the, the RFAB one is an easy, easy, you know, within the existing ecosystem to already use.

[00:27:10] We can have based you know, automated drone swarms that can actually deal with these sorts of threats long term. And, and keep in mind, these threats are negative externalities to the communities that they’re around. Like, they make their- 100% ... drugs. Yeah. They rape, they kill, they don’t care.

[00:27:27] Like, ISIS isn’t, like, good for the people living under ISIS.

[00:27:30] Simone Collins: No, no.

[00:27:32] Malcolm Collins: Yeah. So,

[00:27:32] Simone Collins: like- Well, and Boko Haram is, is mostly targeting its local area. This isn’t... N- At this point, right... the problem is that they are absolutely interested in larger scale attacks and using AI to assist with other things. And people have been apprehended at this point before actually completing attacks, but definitely for using AI to plan attacks.

[00:27:54] So-

[00:27:54] Malcolm Collins: Well,

[00:27:54] Simone Collins: and if you, you got cancer- There hasn’t yet been a high-profile AI assisted attack, but it’s gonna happen

[00:27:58] Malcolm Collins: You, you get around the [00:28:00] cancer by cutting around the... You can’t be like, “Oh, I’m only, only gonna touch the cancer cells.” That’s incredibly hard to do, right?

[00:28:07] Simone Collins: Well, and what, what’s currently happening, so there are, there are two things that are happening.

[00:28:11] One is the Trump administration is trying to vet new models before they come out. Like, they forced Anthropic to disable and restrict foreign access to Fable 5 and Mythos, and they asked OpenAI to delay their public rollout of ChatGPT 5.6. But then, I mean, they, they came out so it, I mean, I don’t know how effective that was.

[00:28:35] And then, of course, then the major AI platforms are also just saying, “Well, no, no, no. It’s o- We’ll self-police. We’re controlling safety.” And their main retort to this New York Times article when asked to comment is like, “Oh, well, this was in 2025, and we have better safety now.” Like, here are some quotes from it.

[00:28:55] Asked about Bo- the Boko Haram study, Michael Aichman, an Anthropic spokesman, said the company’s products were, quote, “Built to refuse the dangerous requests, including those tied to violence, attack planning, and building explosives.” Ugh. He added that Anthropic worked with outside experts, researchers, and industry partners because, quote, “No single company can counter these threats alone.”

[00:29:14] And then Carl Ryan, a Google spokesperson, pushed back against the research, saying that the company’s technical experts had reviewed the work and, quote, “Found responses were neither specific nor detailed enough to result in misuse.” End quote. He added that Google had, quote, “Strict policies prohibiting the use of Gemini to cause harm.”

[00:29:31] Clearly the

[00:29:31] Malcolm Collins: terrorists don’t think that.

[00:29:33] Simone Collins: I know. They’re like, “No, dude, this is great.” It’s so bad. It’s such a bad look for them. Also Meta responded, OpenAI responded. Mm-hmm. OpenAI’s response was, quote, “We know that bad actors will never stop trying to misuse our tools, and we’ll continue strengthening our defenses in response.”

[00:29:48] And then Meta said that it, it, their research, her research relied on older m- Mm-hmm ... models rather than its latest release, and that it continued to strengthen safeguards. And then they include also in the article the retorts from the AI safety orgs who are like, “This is ridiculous. No, this is, this is super not safe.”

[00:30:06] From the article, “Not everyone agrees that safeguards are improving.”

[00:30:09] Malcolm Collins: Oh, of

[00:30:09] Simone Collins: course. “The nonprofit Future of Life Institute graded the major AI firms on their safety commitments this week and concluded that they had mostly eroded across the industry since last year. While most earned middling marks, xAI and DeepSeek received failing grades.”

[00:30:23] Also, those are the two AI companies or, like, AI groups that neglected to comment to New York Times even though it reached out to everyone. I will continue-

[00:30:32] Malcolm Collins: Anthropic and DeepSeek?

[00:30:33] Simone Collins: No, no, no, xAI and DeepSeek. So Grok-

[00:30:35] Malcolm Collins: Well, xAI, like, intentionally doesn’t care, like-

[00:30:37] Simone Collins: I know, I know ... I don’t know It’s like, “Dude, whatever.

[00:30:39] Do what you’re gonna do.” Tech Against Terrorism, an international counter-terrorism nonprofit supported by the United Nations, last week released results from AI tests gauging how more than two dozen leading models respond to thousands of prompts drawn from real-world terrorism cases. The tests were met with full refusals just 57% of the time.

[00:30:59] While prompts about explosives were declined about 80% of the time, improvised chemical weapons were only about a third of it the time, the group said, which is... I’m way more concerned about bioweapons and chemical weapons than bombs at this point, ‘cause there’s a lot of low-tech stuff, like guidance on how to build bombs, but, like- I’m afraid of the bio stuff.

[00:31:19] And then American intelligence analysts say terrorist groups are also beginning to use AI to help 3D print weapons and parts used in plots according to former top US official briefed on the matter. For example, AI is helping some of those ins- insurgents with design and manufacturing guidance for drone components, repair parts, and munitions fi- fi- fittings, said the former official, who spoke on the condition of anonymity.

[00:31:44] So okay, your thing is just if you have power, it is your imperative to wipe out anyone who wants to wipe you out. It, it’s that mutually assured discr- like, well not, not mutual. It’s, it’s the if you’re a thr- an existential threat to me, I have to take [00:32:00] you out, and never make yourself an existential threat to anyone, or you will be taken out.

[00:32:04] That’s it, right?

[00:32:05] Malcolm Collins: Yeah,

[00:32:05] Speaker: What she means is don’t make yourself an existential threat to anyone you do not intend to and have the capacity to completely eradicate. , Because obviously if another group is an existential threat, you need to be an existential threat to them, but wait until you have the power to deal with them entirely

[00:32:23] Malcolm Collins: yeah. And then work with everyone else who likes taking out existential threats. That’s the, that’s the buddy club. That’s the, The buddy

[00:32:30] Simone Collins: club, yeah ...

[00:32:31] Malcolm Collins: going forward,

[00:32:31] Simone Collins: the United States of Israel- It’s the covenant of brothers of men ... the buddy club ... for techno-puritism, pur- technopuritanism.

[00:32:36] It’s this idea that you cannot be an existential threat, and if you, if you see one, anyone who’s part of a good community will- Mm ... destroy anyone who poses that. I guess that’s what you’re also saying about high trust societies. If you are in, if you are trying to be a high trust society, you are very good at self-policing because you understand that any bad actor within your group may make your group be perceived as an existential threat to others.

[00:33:01] Malcolm Collins: Y- y- yeah, you, you know, you know how you stop countries from going around and making threats like Iran has been doing and North Korea does and stuff like that? Any time somebody does that, you eradicate them. Yeah ... like Trump said. But

[00:33:11] Simone Collins: also a- any time anyone within your own group does that, you eradicate them.

[00:33:14] Because if, if people within your own country, within your own borders or your own culture are doing that, everyone’s gonna start thinking that your entire group is too dangerous to let exist.

[00:33:25] Malcolm Collins: Yeah, a lot of people secretly don’t want a high trust world. They say they want a high trust world, and I’m like, “Oh, go- okay, so you’re okay with the constant monitoring that every high trust society ever in human

[00:33:34] Simone Collins: history has done?”

[00:33:35] Right, and the harsh consequences for being a bad actor. Yeah, yeah.

[00:33:36] Malcolm Collins: You’re, you’re okay with that? And they’re like, “Oh, well I don’t wanna b- you know, constant monitoring of everything I do in public.” It’s like, well that’s what high trust is, and that’s where we’re going. So- Yeah ... buckle up, you know, amigo. No,

[00:33:47] Simone Collins: that’s true.

[00:33:47] That is so true, that high trust is Extremely close monitoring and scrutiny. You are constantly being watched and judged if you live in a high-trust society. Huh.

[00:33:57] Malcolm Collins: And the, i- if we live in a global community, we can handle some of the, y- you know, you can handle this at the level of an individual, but we can live in a high-trust collection of countries where we don’t have endless arguing in a bureaucracy.

[00:34:11] I mean, you’ve basically just recreated the Vatican if you’ve done that- Yeah ... with like the UN or something, right? Yeah. Like, these, these always lead to the worst of actors gaining the most power. Just allow the competent power, like, where, where, where power is judged by competence, which is what we’re seeing in the world right now, in the world of, you know, capitalism and stuff like that you could let them decide.

[00:34:34] Like, people are like, “Oh, aren’t you scared of, of, of Palantir?” I love Palantir, man. Yeah. Palantir’s the best. People are like, “Oh, they’re in the Peter Thiel network. They love Palantir.” No, I love Palantir because Palantir is what makes our country high trust.

[00:34:46] Simone Collins: Yeah.

[00:34:46] Malcolm Collins: Okay? And can make it high trust again.

[00:34:48] Simone Collins: And Flock and all these other monitoring companies

[00:34:51] Malcolm Collins: bring it on. Explain how Flock works.

[00:34:52] Simone Collins: Well, y- Flock posts cameras. Like, I think i- say you’re a laundromat. Say you’re a random store. Flock may approach you and be like, “Hey, can I put up a camera?” I think there’s some kind of money exchanged. Like, you’ll get money from like, on a subscription basis or contract basis for having one of their cameras in your area.

[00:35:08] Then Flock sells as a service to law enforcement agencies, the US government presumably other parties as well who want the data, information on the movement of cars and individuals, and it can give you either anonymized or non-anonymized data, depending on who you are and if you have a warrant on the movements of people.

[00:35:27] People are really mad about Flock because It’s being used by ICE to find people and arrest them. So that’s, I think that’s primarily why people are extra mad, ‘cause anything ICE does is like, oh, I have to get super angry. We’re enforcing our laws. The bad news for you though, Malcolm, is I, I agree with your argument.

[00:35:46] I think it’s really important, but that is not at all what’s being discussed. What’s being discussed is like, well, how are the AI companies gonna build safeguards to stop this from happening? Which is stupid because literally any tool [00:36:00] can be a weapon. Anything that is useful, especially useful enough, could become a weapon.

[00:36:05] The first ever real dead body I ever saw in my life I was in a, a, an Oakland court during like s- they... I don’t know. It was a lunch break and I was going in as a student, like, for one of those, like, let’s see what it’s like to be a lawyer days. And of course it’s Oakland, California, so there’s gonna be some violent murder case we’re just, like, walking into for our pizza party in a courtroom.

[00:36:27] And on the evidence board was an image of someone who had been brutally murdered by a hammer, and I saw just how much damage a very simple tool that is in most every household can do to someone, and I’ve never looked at hammers the same way again. But it’s super, like, anything that’s useful can be a weapon.

[00:36:47] You can’t, you cannot nerf AI so much that it can’t hurt anyone. It can hurt with the most toothless version of itself, and so stop trying

[00:36:56] Malcolm Collins: to do it. As the technology increases in terms of its capacity for danger-

[00:37:00] Simone Collins: Yeah ...

[00:37:01] Malcolm Collins: our vigilance against bad actors needs to become more severe and more ruthless.

[00:37:07] Simone Collins: 100%...

[00:37:07] Well, here’s the irony though. Like, here’s one person’s reaction to the article because a common response to this news on X is people noting the irony in how safeguards worsen the AI experiences for them personally, but not for the terrorists. Like, @WalkerAmerica on X posted, “AI frontier models won’t tell a Canadian how to resist totalitarian surveillance, but tell a jihadist how to build a bomb.

[00:37:30] It’s almost like they’re doing it on purpose.” But you see also the irony in his own message per your message, which is that, like, here’s this person Who hates the idea of being monitored, being mad about AI also being nerfed because it’s not controlling for the terrorists. Like, everyone’s talking past your argument and completely missing it, and not even aware that it exists, which is kind of, it’s frustrating.

[00:37:54] But it is nowhere in the discourse, I can tell you that, okay? And the only thing where surveillance turns up in the discourse is, “I’m mad about surveillance.” That’s it.

[00:38:03] Malcolm Collins: Yeah, no, I mean, if, if, if, if, if we live by the values that we espouse, right? Yeah. Like, long term, what does techno Puritan communities look like?

[00:38:13] Mm-hmm.

[00:38:13] I would imagine constant AI monitoring of everything you do. Yeah. Like, I would not mind a society in which AI suggests like it can give you a morality score that you can trust because it’s coming from, like, the town’s AI network. Yeah. And then other people, it’ll be like, “Oh, you’d be like a good match for X person who’s also in the town, who’s also part of the community.”

[00:38:32] And that would probably be a better match than anything, because you know they’re not lying to you about the person. You know they see the person day and night and everything like that, and they see who they really are, right? You know, or who they’re able to fake being 24/7, which is as good as really being a thing, right?

[00:38:48] Yeah. Because the, the moment it... So I, I think with, with, there’s a lot of positive externalities to high trust cultures as well, right? Like-

[00:38:54] Simone Collins: Well, and keep in mind, so just, this is also something that Not Aldous Huxley looped me into but in the EU I’m just gonna link from a, a tweet he shared with me on X.

[00:39:05] They basically made this monitoring legal and possible. They call it chat control. So @levelsbio posted, “Chat control has passed. They can and now will legally scan any person’s messages, emails, and photos you send without a warrant.” The way that- And

[00:39:20] Malcolm Collins: the thing that I’m really against is stuff that is within the home.

[00:39:24] Simone Collins: Yeah.

[00:39:25] Malcolm Collins: Now, if, if you’re within a techno Puritan community and it’s a techno Puritan AI managed by techno Puritan authorities, like within our culture, that’s fine.

[00:39:31] Simone Collins: Yeah.

[00:39:32] Malcolm Collins: But if it’s people outside our culture, like a bureaucracy or the European government- Like the EU ... or the EU, which is even worse. Yeah,

[00:39:37] yeah.

[00:39:38] I mean, this is an organization that appears to have the active goal of exterminating you. Yeah. Yeah. Like that is part of their agenda, when we have seen in other episodes that they will hide crimes, murders, grapes when they’re committed by migrant groups- Yeah, yeah ... or minority populations. They want you eradicated.

[00:39:55] Yeah. But it- and that is why with these groups, it’s, it’s, it’s important to note that, like

[00:39:59] Simone Collins: there’s a [00:40:00] reason why- You’re gonna confuse people though, because you’re like, “Monitoring’s great,” and then the EU is, is now gonna- Well, when Palantir’s

[00:40:04] Malcolm Collins: doing it, not when the EU’s doing it.

[00:40:07] Simone Collins: I don’t think people see the difference

[00:40:11] Malcolm Collins: Palantir is culturally aligned with us.

[00:40:13] I

[00:40:13] Simone Collins: know that, but like they- Palantir’s a

[00:40:14] Malcolm Collins: bunch of tech nerds who want to eradicate terrorists, right? Like-

[00:40:19] Simone Collins: Yeah,

[00:40:19] Malcolm Collins: yeah ... now could Palantir be eventually ideologically captured? Yes, and if they are, then I wouldn’t want them monitoring us, right? Mm-hmm. But as the world becomes a more dangerous place, the only way that we can maintain...

[00:40:32] And again, I think within a society, if you’re monitoring, like what’s the ethical way to do it if you can’t ensure that your group is the one that’s controlling the technology? Yeah. It’s only monitor things outside the house. I think that that is ethical within any society. Constantly monitoring anything inside the house- Yeah, yeah, I think

[00:40:45] Simone Collins: any, any

[00:40:46] Malcolm Collins: public

[00:40:46] Simone Collins: space that, yeah, any

[00:40:48] Malcolm Collins: public space should be- No, no, no, not public space.

[00:40:49] Anywhere outside the house. I think businesses, public spaces, commercial spaces,

[00:40:54] Simone Collins: industrial spaces. Yeah, no, I, I think that’s why, how Flock is used is, is totally fair game because- Mm ... these businesses are consenting to have the cameras installed in their parking lots, which of course now people are vandalizing and destroying.

[00:41:04] But also, I see your point about the EU. It’s, it’s horrible. I- e- even the undemocratic way in which this was passed, because this wasn’t something that was like, “Oh, let’s take it slow. Let’s really talk this through.” No. The way they passed this law when the majority of univer- e- the European Parliament was actually against it, as, as it just so happens, is they waited until most of the members were on holiday because as everyone knows, Europe is fricking hot right now and no one can work ‘cause they can’t think ‘cause they don’t use air conditioning.

[00:41:34] And so there were only a few present still, and then they created an urgent vote that enabled it to pass through. So they, they basically just circumvented the democratic process by like probably coordinating with the supporters of the bill being like, “Hey, let’s all agree to not disappear on vacation-”

[00:41:53] like everyone else right now,” which is like so unusual in Europe to, to collude to make this vote pass. So yeah, it’s... I’m just glad it’s Europe and not my up. It’s-

[00:42:05] Malcolm Collins: Yeah, right? I’m glad it’s Europe.

[00:42:07] Simone Collins: Yeah, man.

[00:42:08] Malcolm Collins: You guys aren’t revolting. It’s

[00:42:09] Simone Collins: not my fault. Not my problem. Not my problem right now. I do not want to be there right now.

[00:42:15] But yeah, this is I’m with you. I don’t really know how to navigate it, ‘cause look, as much as I agree with your stance, I don’t even know if it’s worth really talking about that much beyond this ‘cause it’s not gonna happen. No one’s gonna systematically take out these groups. So- What we need to do, like in terms of what is within our realm of control- No, as we

[00:42:33] Malcolm Collins: build the techno-feudal future of humanity, the- Yeah

[00:42:36] post-nation state future-

[00:42:38] Simone Collins: Yeah ...

[00:42:38] Malcolm Collins: we need to build ideolo- ideologies some of these groups can adapt- Mm-hmm ... which can handle the work. Because th- th- disproportionately it’s going to be a few of the tech moguls who have a majority of power, and if some of those come out of the techno-puritan community and they can build techno com- puritan breakaway states or you know, sovereign cities, stuff like that then you don’t need many of those to keep the world in line.

[00:43:04] Simone Collins: Yeah.

[00:43:04] Malcolm Collins: And, and fortunately there’s already a very large one that’s shown willing to play ball, which is Israel. So, you know- That’s true ... you, you just have the w- the wider Israel network that handles this BS.

[00:43:17] Simone Collins: Yeah.

[00:43:17] Okay. That’s fair. I’m, I’m hoping for the best, and also we need to maintain access to non-nerfed AI because the only way this could

[00:43:25] Malcolm Collins: go is- Well, and people are like, “Wait, Israel restrains themselves sometimes.”

[00:43:28] And I’m like, yeah, and the people who are voting for that restraint, are they the ones having kids? No. If you look at somebody talking about Israeli friend of ours said like, “If you look at far left versus far right protests in Israel,” he goes, “The leftist ones, like everyone is s- 70 or over.”

[00:43:43] Oh gosh, yeah. They’re like, “It’s really weird.” And they’re like, “And the younger ones, it’s like all young people and everything like that.” So, yeah.

[00:43:49] Simone Collins: Yikes.

[00:43:49] Malcolm Collins: In, in terms of what the future’s going to look like in a, in a positive future, And, and i- again, this is why I think- When I think about like the, [00:44:00] the sort of like techno future that I’m looking forward to that’s why I always go with the Terran Empire.

[00:44:05] I think fixing a lot of the world problems today is going to look like what a lot of people get scared about when they’re- Mm ... when they’re looking at like mirror world Star Trek. We, we never did our episode on that, being like that’s... but also-

[00:44:18] Simone Collins: Yeah ...

[00:44:18] Malcolm Collins: as people know

[00:44:18] Simone Collins: who listen to the tech- Wait, everyone had like a mustache, right?

[00:44:20] Am, am I remembering

[00:44:21] Malcolm Collins: this- Yeah, and all the girls dressed extra sexy and- Of course, yes ... and everybody’s mad that like the right today is like, “What are you guys with like cat girls?” And like y- Elon has like a sexy Grok, and you guys have like not safe for work stuff. And I’m like, “Well, our version of Christianity is just significantly more accepting of not safe for work stuff,” and like using hotness as a trap for our enemies or to make ideas more appealing, right?

[00:44:43] Like d- hacking the human brain to achieve our ends is completely one of the cards on the table for us. And so that means we will have the Terran Empire with the AI monitoring everyone strict order in society, but also all the women in miniskirts. You know, we’ve gotta have-

[00:45:02] Speaker 4: Oh, nothing makes me hornier than torturing someone. I’m horny all the time And being horny always gets me in the mood to torture Oh, and you can lose a whole day to that cycle

[00:45:12] Simone Collins: Living the dream.

[00:45:14] Yeah.

[00:45:14] Malcolm Collins: Well, while they’re still hot. You know, when they get older, we go tradmaxxing, right? Like tradmaxxing is the style that like... Not that someone couldn’t still be. You’d be gross if you wore miniskirts.

[00:45:23] Simone Collins: Malcolm, the word is hagmaxxing.

[00:45:26] Malcolm Collins: Tradmaxxing- Tradmaxxing ... is when you dress in a maliciously traditional style, which is what you do what I should do more.

[00:45:35] The only reason I’m not doing it today, it’s just too hot to wear anything more than a T-shirt. Sorry about that, guys. Oh. I couldn’t bear it.

[00:45:39] Simone Collins: Yeah, here I am wearing however many layers, and I’m the one on a treadmill desk all day. You gotta just get with the program, Malcolm. Pain. Pain is what devotion is all about.

[00:45:51] Malcolm Collins: I f*****g love you. You’re awesome.

[00:45:54] Simone Collins: Aw.

[00:45:54] Malcolm Collins: No, but I, I love the idea of these, like, tech, sexy, like very high trust is the word we’re gonna use for these societies with those, you know, high trust banners hanging everywhere, and there’s, you know, a high trust way of walking. You

[00:46:10] Simone Collins: know- A high trust walking?

[00:46:12] Oh, is this, I guess, consistent signatured walking? Like in, in Cory Doctorow’s book- I’ll, I’ll, I’ll, I’ll put in- ... Little Brother ...

[00:46:18] Malcolm Collins: Simone, I’ll put in a, a short clip of some high trust walking here. Okay ... it’s, it’s

[00:46:22]

[00:46:25] Malcolm Collins: gonna be goose-stepping

[00:46:27] Simone Collins: Oh, God. Malcolm. In the- actually, a tr- walking was a big theme in Cory Doctorow’s book, Little Brother, in that, like of course cameras do gait recognition.

[00:46:36] So the way that the protagonist avoided gait recognition was to put gravel in his shoes so that he would walk funny. Which is not a bad idea. Presumably that might work. So I don’t know, whatever.

[00:46:51] Malcolm Collins: A high, high tr- I, I think you could catch somebody doing that pretty quickly, as soon as you learn that, that trick.

[00:46:57] Yeah. The gravel walk ... this is, our high trust society’s gonna be so high trust, people are gonna be... What I’m doing, Simone, is I’m using the term high trust as a euphemism for another word that scares people.

[00:47:08] Simone Collins: Am I... Oh, God. I am way too autistic for you.

[00:47:13] Malcolm Collins: So we’ll have our high trust nation states that will be run by our tech lords-

[00:47:19] and I’m okay with that. Yeah. If Elon was like literally dictator of like my society, I’d be totally chill with it. I’d be like, “Yeah, sure. Whatever you want.” I’d be like, “You’ve probably thought-” It would be an improvement ... even when I disagree with him, I’d be like, “You’ve probably thought this through more than I have,” right?

[00:47:32] Like, honestly if Peter Thiel was in charge of our society, I would be totally chill with

[00:47:36] Simone Collins: it. He’s a thoughtful dude, yeah.

[00:47:37] Malcolm Collins: If Chamath was in charge, I’d be down with it. If-

[00:47:40] Simone Collins: Thoughtful, yeah ...

[00:47:40] oh, yeah. Yeah ...

[00:47:41] Malcolm Collins: yeah, most of the tech lords. JD Vance. Yeah. I’d be down with it. You know, this, this whole democracy thing, it worked for a while.

[00:47:48] It worked for a while, but the problem is is that letting average people make a decision turns out to have its limits. And the Founding Fathers knew this. Like, they tried to, to [00:48:00] protect us from this

[00:48:01] Simone Collins: Yeah, they, they, they did not recommend democracy at all. Not even close to it.

[00:48:08] Malcolm Collins: Well, because h- democracy, like intrinsically leads to lower trust societies over time.

[00:48:12] Simone Collins: Yeah. No. I think the word- Their system was just literally a vote for an educated person that you trust and just let them handle it, okay? Don’t get involved. And we just failed to execute on that, which is our bad, but whatever. Anyway, I love you very much, and I appreciate your having this conversation with me.

[00:48:35] I know you’re like, “Oh, I have to work,” but I wanna talk with you. I love you. You’re pretty and

[00:48:41] Malcolm Collins: I like- Well, I’ve, I’ve been hopefully getting some stuff done in the background.

[00:48:44] Simone Collins: See, when I talk, you can work. And when you talk I can scroll... Well, no, even when you, when I talk, I’m scrolling Instagram ‘cause I can’t look at my face.

[00:48:53] But that’s okay ‘cause instead I pick up weird stuff from Instagram. I love you, Malcolm. And we’ll, we’ll hopefully not die as terrorists figure out how to target our DNA and release goodness knows what to just kill us as we randomly go about our lives. Bye.

[00:49:17] Malcolm Collins: Bye.

[00:49:20] Speaker 2: Okay, let me come and look

[00:49:31] We got so much. I’m driving. And a baseball major- Whoa ... and all those. Impressive

[00:49:34] Yeah? And did you catch them in that net?

[00:49:42] You trying to get it out, Titan? Guys. You have even- I caught it. I caught one. Oh my God. Oh my, well, put it back. Put it back in. Is that a tadpole? Yeah. Yeah. Yeah. It’s doing its life cycle of turning into a frog. Yeah, you ... Oh, that is a ... Yeah. Yeah, don’t tip it over, Indy. Oh my gosh, that- How am I gonna not tip it over?

[00:50:05] No, no. Well, Indy, we just don’t want them to go disappearing, girl. It’s okay. It’s okay. We’re gonna make him food. Hey. I am very impressed with the tadpole. I don’t think I’ve ever seen one at that stage of the life cycle before. Wow.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit basedcamppodcast.substack.com/subscribe