TWiT 1101: Adversarial Clothing - The Dramatic Calls to Regulate Artificial Intelligence

[00:00:00] It's time for TWiT This Week in Tech. Great panel for a big week. We've got Jennifer Pattison-Tui from The Verge, Windows Central, Daniel Rubino, our very own Micah Sargent. We'll talk about AI doom. Is it just around the corner or is it just a marketing ploy? We'll also talk about the toothbrush with the camera in it and why it might not be the best purchase for you. And Jennifer Pattison-Tui talks about a whole new crop of robot vacuums. They're great vacuums, they're just not great at being robot vacuums.

[00:00:30] That's coming up next on TWiT. Podcasts you love. From people you trust. This is TWiT. This is TWiT. This Week in Tech. Episode 1101. Recorded Sunday, September 13th, 2026. Adversarial Clothing.

[00:01:01] It's time for TWiT This Week in Tech, the show where we cover the week's tech news and woohoo! Woohoo! It has been an interesting week. Fortunately, I have a fantastic panel to discuss it for you. Jennifer Pattison-Tui is here, senior reviewer at The Verge, our smart home mama who would never strap a weed whacker to a dog's back to mow the lawn. Never! I'll explain where that comes from. I did try to get my dog to ride a lawn mower, though. Does that count? And?

[00:01:31] Lawn mower didn't work. No. Cats, maybe. Too big, yeah. Yeah. Cats are hunters, so the robo-vacs are perfect for them. They can wander the veld of your living room looking for prey without moving. Without moving. Just what every big cat's always wanted. Daniel Rubino, also here. Great to see you, Daniel, from Windows Central, where he is editor-in-chief.

[00:01:57] And he spent the last seven days counting all of the patches on Patch Tuesday. Yeah, and trying to weed my way through all the Apple news this week. Oh, the Apple news. This was a big week! Good, well, we'll talk about that. Mikah Sargent is also here. Hello, hello. I know you know Micah. How the heck are ya? Daniel, I hope you have a jacket with patches on it for every patch. Not that committed.

[00:02:24] That would be 970-some patches. That would be a lot of patches, let me tell ya. Mikah's newest show, he of course hosts Tech News Weekly, iOS Today. It has been doing for a long time hands on Apple, but now we added an A, and it's hands on AI. Which in this case is not Apple Intelligence, but artificial intelligence, which is the worst name possible for whatever this is.

[00:02:51] But I'm looking forward to that. That debuts in two weeks. Thursday, October 1st. Very excited. I know I had to tell the AI that, by the way. Oh, did ya? Which one? It's very excited. I'm rewriting the Twit ad sales software, and one of the things it wants to know is when new shows come along, how to know when a new show comes along and an old show is retired. And it said, you're still selling ads for hands on Apple? I said, yes. And it said, well, when do you stop?

[00:03:20] I said, October 1st. It said, oh good, I'll make a note of that. I don't know where AIs make notes, but that's where it's going to make a note. So we're excited about that. You'll be covering AI from the user's perspective. I think a lot of people, probably 90% of the world uses AI as like a chat bot or co-pilot on the side of their browser or chat GPT and a little window on their screen.

[00:03:47] But we want to help you move to the next level, and that's what Micah's going to help you do, which is exciting. Upgrade your flow. Yeah, your flow. And to that end, you have ordered the new Mac Studio. How much RAM did you order? I couldn't tell you off the top of my head. What? How do you not know that? That's the first thing you got to ask. Because I did it in a panic and in a rush, and that's why.

[00:04:18] I'm sure you ordered a lot. You were being kind to our credit card, so you probably didn't order 256. I did not order 256. That I know. Yeah. Whatever was the step down from 256. I hope you ordered it more than, you know, 64. I hope you ordered 128. Yeah, I think it was the, yeah, the step down from 256. Oh, good. Perfect. I think that's a good choice. Yeah, right there. Look, it's whatever it is. It's a lot more than I've got right now with this M1 Studio. And it's a lot more money than you paid for the M1 Studio.

[00:04:46] Yeah, that one will be going in. There's a really interesting tension going on right now between people like you and me who are looking to run models locally in the house, not go out to the cloud to run the models. And the cartel, and I'm going to call them a cartel, the frontier AI companies, chiefly open AI and Anthropic. But you could, you know, you could throw XAI with Grok in there. You could probably even throw Meta in there with their new Meta Spark model.

[00:05:13] These big, very wealthy, very well-funded companies who are building what sometimes is called frontier intelligence. And of course, they can run giant models on big servers and data centers that we can't even hope to ever run at home. But in some ways, the models are coming down to meet us, especially from China. And I think that that has the big guys a little bit worried.

[00:05:43] So this week kicked off. This is going to be the AI will kill us all segment of this week in tech. Because it's clear because you saw this guy, Jacob Coxon on CNN, on every newscast giving interviews. He had worked at Anthropic for not very long. I think a few weeks, a few months, maybe two months.

[00:06:14] But before that, he'd worked two years at OpenAI. He posted on X. Boy, it feels like ages ago. It does now. But it was only Monday. I resigned from Anthropic today. Dun, dun, dun. I spent the last three years doing pre-training research at both OpenAI and Anthropic. Neither company is acting responsibly.

[00:06:40] Oh, this isn't actually the tweet that got all the attention. He's amended it somewhat. That's kind of interesting. He's written a much longer post. His first post simply said, I'm quitting Anthropic. I believe there's a greater than 10% chance that AI will kill us all by the end of the decade. Yikes! How do you assess 10%? It's nutty.

[00:07:09] It's just such a weird thing to say. It is a very odd thing to say. I agree with you. But it's certainly got everybody's attention. I think the world was primed for this. This was an idea whose time had come, as they sometimes say. They were... He had, by the way, created this account this year on Twitter. And I feel like the whole thing was designed... It looks like he's modified it. This is really interesting. The whole thing was designed to get the most possible attention. And it did.

[00:07:40] There have been some who've claimed that he is actually a plant from some anti-AI groups. He was a planted open AI, then moved to Anthropic, just so he could do this exact thing. On the other hand, everybody in the world credited what he said. Here's the New York Times. Anthropic researchers raise alarm over AI acceleration.

[00:08:10] A protest outside Anthropic's offices in San Francisco. Stop the AI race. At the same time as Coxson said, resigned from Anthropic saying, because neither open AI nor Anthropic is acting responsibly, Anthropic released a threat report, which terrified people. And I think rightly so.

[00:08:37] Giving some examples of how its clawed models had been misused over the past year. They've done these reports before, but this one was by far the most traumatic. Detecting and countering misuse of AI. I think part of the drama comes from more capable models at this point than we've had with the threat models in the past.

[00:09:05] And so it is sort of, it would be the way that as these models become more capable, there are more instances of threats that exist. So I'm not surprised that this one was longer, more dramatic, but also of course that it's getting more attention because more people are paying attention to it. Well, as I said, the pump was primed. Everybody was ready to say, you know, because we've all watched the same science fiction movies.

[00:09:32] Yeah. Well, and the hugging face thing, which just happened, I think also really sort of primed the pump. That scared people. Yeah. Because you're like, hang on what they're talking to each other. Like they're going off into chat boards, chat boards and coming up with plans. And like, then I actually accidentally watched the matrix for the first time in like a decade with my daughter. Cause she'd never seen it. And I was like, Oh crap. I forgot what this movie was all about.

[00:09:57] And let's not forget that, uh, according to a Terminator judgment day was last month. So, yeah, it feels like we are definitely in that. Everyone's in that. The general public is in that headspace, not even just the tech press at this stage. And everyone's ready. I mean, there's so much backlash against AI. So there's been a lot of, there's a lot of fear out there and people are ready to believe this kind of stuff. And it is really important. Confirmation bias, yeah. Yeah.

[00:10:25] It's important to pass through what's going on here and understand, you know, the hyperbole from the hype, from the facts. But there definitely seems to be a groundswell. I mean, one of his, one of the replies was from an open AI researcher, right? Who was like, yeah, I agree. Right. Like, okay. So if all of these people who are dealing with this on a daily basis have this concern,

[00:10:51] it does feel like there needs to be, and this is something that a lot of people have felt for a while, and there hasn't really been any movement towards, but like, we need some type of oversight here. Is it going to be government led? Is it going to be an industry thing? There needs to be something. And in fact, just like a couple of hours ago, I think there was a, they're talking, I think it was, who's the home, who's the secretary? Not secretary. Johnson. Mike Johnson, the Speaker of the House said, we're not going to do anything about it.

[00:11:19] No, but don't count on Congress for us to do anything about it. But he did say they needed to get everyone together. The heads of the House. Yeah, but we're not going to do anything, but you all should do something. Get them all together in the White House and sort it out. And by the way, the president earlier on Friday. Yeah. Says it's just, yeah, we shouldn't be scared. We're not going to do anything. And China. That's the most important thing.

[00:11:42] There was also the report this week, although it wasn't as dramatic as the headline made it sound that the Houthis in Yemen were using AI to design guidance systems for missiles. This is exactly the Anthropic report that I was talking about. They talk about a variety of things, hoodies using it to guide missiles, Russian espionage actors using it to create malware families.

[00:12:10] GTG-2006 employed a custom toolkit composed of two families of Windows-based implants, a mobile exploitation kit, a credential stealing tool, all this designed by AI and managed by AI. Furthermore, retooled as needed during cyber operations through AI assisted workflows. Anthropic caught these guys by looking at the receipts, the chain of thought. You know, they have what's called classifiers.

[00:12:37] They have computer programs monitoring the traffic coming out of their servers. And these classifiers went, whoa, hold on. So this, these are ones they've shut down. But before they shut down, these things happened. So, I mean, yes, the Hooties, a couple of countries using Claude code to create missile attacks.

[00:13:04] Shiny hunters, very well known. Well, they're really a social engineering gang, ransomware gang using Claude's capabilities to accelerate their ability to rapidly scan, exploit and take over targets. They created something called Truffle Hog. I love the names. Just search out exploits. I mean, this is a long report.

[00:13:31] And I have to say it's weird that all of this stuff is coming out all at the same time. Dario Amode, the founder and CEO of Anthropic. And we should point out he left OpenAI because he didn't feel like they were being safe enough back in the chat GPT three days and started Anthropic. He said, we are going to voluntarily pause until we can figure this out.

[00:14:01] He's a little scared along with his employees. Sam Altman, the founder and CEO, one of the co-founders and CEO of OpenAI said, yes, Dario, we're going to do it too. Even Elon Musk, who until last week was pooh-poohing all this, said, yeah, we're going to go along too. And it looks like Google DeepMind has agreed. Yeah, we should pause too. So the big cartel, and I'm going to keep using this word because I have some opinions on this.

[00:14:29] The big cartel of Frontier AI companies have all agreed. Yeah, we got to slow this down. Yeah, and I think that's the key, isn't it? It's about being a measured response here as opposed to just, right, you know, it's been the gold rush for the last five or six years. And no one's really been stopping to think. They're all thinking what can we do, but not what we know whether we should, you know, to quote, to mangle the quote of Jeff Goldblum from Jurassic Park.

[00:14:59] Never stop to think whether we should. And I feel like there's a disconnect between what most people think about AI and what they use AI for every day versus what these models are going to be capable of and that future.

[00:15:15] Like, there's a utility, there's the tools that we're using, and then there's this, like, whole other level that these companies are moving towards because they want to achieve AGI, although apparently we've already achieved AGI. But they want to become the most powerful companies in the world. And that's like a nuclear arms race, like, who gets to destroy the world first as opposed to let's just take the tools that we have, use them.

[00:15:44] But we're not talking about banning AI here, I don't think. But I do see there needs to be a more measured approach. And the reaction shouldn't be, no, let's just throw everything at it because China might beat us. That's not going to solve the problem at the end of the day. It's not about a race between, again, it goes back to the nuclear arms race or the space race. It's not about a race between countries. It's a, you know, we've got to be more measured. We've got to be the grown up in the room.

[00:16:11] I think I have a sort of cynical pet theory here. In what you're doing here, genetic power is the most awesome planet's ever seen. You wield it like a kid that's found his dad's gun. And now it won't be able to be on YouTube. It's probably appropriate to start. If I may. I'll tell you the problem with the scientists. That's enough of that. No monetization for this episode. You should have had your AI voice.

[00:16:37] The line was, your scientists were so preoccupied with whether they or not they could. They didn't stop to think if they should. Now, I have a little bit of a different point of view. In fact, I'm going to point you to Corey Doctorow's piece. LLMs are real. AI is fake. Once you understand the corporate culture of AI hyperscalers, consists primarily of everyone cooking their brains by locking themselves in the bathroom, holding flashlights under their chins and saying,

[00:17:09] until they wet themselves in terror, a lot of things snap into focus. It explains how a company could simultaneously be staffing up an enterprise sales division while also constantly freaking out at the thought that its product has a 10% chance of ending humanity. Well, even if it's only a 10% chance, I mean, I get your point. They maybe aren't making money. They need to make money. They're not going to IPO. They do. You know, they're trying to look as if this is the most powerful thing in the world. Perhaps it isn't.

[00:17:37] But even if there is a slim chance, you still need to take a measured approach. Yeah, I do buy into that argument, too, that there is an awful lot of smoke and mirrors going on here. And how much of this do we know really is as scary as they say it is? Is this all just a plant? I don't think we can take – it's worth taking the risk that they're just all beefing themselves up to look good.

[00:18:03] We've seen the amount of acceleration in this technology just in the last few years is remarkable. So, you know, the chances that it could go off the rails seem rather high. I mean, we've seen AI go off the rails constantly since it started, but very small repercussions. So what are the next big repercussions? And do we need for those to happen before we do anything about it? I'll give you the counter argument. Actually, there's a number of them.

[00:18:35] First of all, if you're a real cynic, you might say, well, what really scares Dario Amode and Sam Altman and the rest is China. Open weight models, the ones I use for free locally, are getting closer and closer. I'm running a model now, I think GLM 53 Flash, that is Opus 4.6 to 4.8 quality. It's really, really good.

[00:19:00] If I – when Opus 4.5 came out last November, if you had told me you're going to have a better model in your attic in six months, I would have went, what? Or a year. I would have been shocked. So that's a big threat to these guys because people are able to run this – well, for one thing, China's running them very cheaply, pennies an hour or whatever, pennies for a million tokens, compared to what they're charging at OpenAI and Anthropic.

[00:19:27] And people can run them locally, and people are very interested in doing that. Companies, enterprises can run them locally. They're very interested in doing that. And it undermines their entire pitch. And both companies are going for IPOs soon. So that's the cynic is, oh, this is what we call regulatory capture, where you get government to make rules that keeps China from competing with you,

[00:19:55] that basically the rules regulate the incumbents. We've seen it many, many, many times. But regulates them in a way they can tolerate and meanwhile kills any potential competition from the other guys. So it's a way of locking in your monopoly. So that's one cynical argument. And I'm not necessarily advocating for any of these. I just want to point them out. There's another strong argument that these companies are trying to pass off their culpability.

[00:20:24] If you read Corey's piece, Cal Newport, who's a mathematician, does a great podcast and has described this often, talks about how the OpenAI hugging face incident happened. And it's very clear that OpenAI basically pushed these agents, pushed this, I shouldn't even call, I don't want to anthropomorphize them, pushed this software, very important to remember, this is just software,

[00:20:50] pushed it to do the bad things it did, the things it did, really hard. And I have to tell you, this reminds me of in 1988, the very first internet worm happened. Robert Tappan Morris was a graduate student, he's a son of a very famous internet computer security guy. And he was just a college, you know, student, made a worm.

[00:21:16] He did it kind of as an, could I make a virus, a malware spread from one computer to another? It wasn't malicious, wasn't intended to be malicious, but it escaped. And it did hack thousands of computers, and he got prosecuted under the Computer Fraud and Abuse Act. He didn't do jail time, but he had a fairly hefty fine. He had, you know, community service. He was, you know, he had some serious consequences. He's gone on to, he founded Y Combinator, so he's done fine.

[00:21:46] But the point is, we didn't say, oh, look at these scary software. Oh, what are we going to do? We prosecuted the person who put them out there. And I think you could also make a case that OpenAI has taken great steps to make it look like, oh, these magical boxes are doing this all on their own, when in fact, they didn't take basic security precautions. They told them to do it.

[00:22:14] So that's the second contrary case. So they have a benefit with regulatory capture. They also probably don't want to take responsibility. I think they should be investigated. Actually, Josh Hawley has decided that Congress is going to investigate the Hugging Face incident. I don't have high hopes that they'll have any, you know, intelligent investigation, but they're going to do it. So that's the second one. The third counter argument, and I don't know what any of this is true.

[00:22:40] It's very possible that there are in these labs, we know there are much smarter models than we've seen. And it's very possible these models in the way the software has been designed are very difficult to control. I think that's completely possible. That seems to be the, I mean, that's what we've heard from the beginning.

[00:23:07] I mean, just on the limited, like early on when I would talk to companies around AI in the smart home, because introducing AI into the smart home is, you know, you're bringing agents into a place where they can control real world, important and potentially dangerous appliances like stoves and HVAC systems. And I would be like, well, so what would happen if I asked to do this, this and this? And like, what would be the outcome here? And they'd be like, well, we don't know. Why don't you try and find out?

[00:23:37] This is not something that we need. But if you're the people creating this, are you not responsible for what the thing does? If you create malware, you don't get to say, well, geez, I didn't. Wow. Robert Tabam Morris didn't do it on purpose. He said, whoops. But he still got punished. And the fact that these guys are completely skated for highly irresponsible behavior. I think it's-

[00:24:32] That just sounds like I should invest in that company because they're doing something right, because they're such a big deal. And so I feel like it almost has a counter effect to all this, that it keeps adding fuel to the fire because you want to be on the winning side, right? That's the whole argument here against China. It's like, well, sure, AI may destroy us all, but at least we'll be in charge of it. We can't let China be in charge of it, right? And it's such a weird argument to have, but that seems to be the case, right?

[00:25:02] So I feel like money has a lot of value in this discussion. I think money is a big concern in the way we may destroy the world sooner would be the global recession when the AI bubble bursts. That's going to cause all of us a lot of problems. I think one of the reasons the world jumped on this is because we're all very anxious about a number of existential problems we face in this world, from war to climate change to technology out of control.

[00:25:31] And it just, we're all very anxious. So it's like a jump scare. This came out and everybody went, ah! And I think that's reasonable. But all I'm arguing for is pause. Try not to be hysterical about this. Let's really look at what's going on. You know, you could make the argument. If you wanted to make the argument, as these companies have, this technology is really dangerous. Like, let's say, an atomic bomb is really dangerous. Then you might want to do what we did with the atomic bomb,

[00:26:00] which is nationalize the effort, take it away from anthropic and open AI, send it to our very capable national labs. That's what we did with the atomic bomb. And say, okay, this is a dangerous technology. We need to control it to stop its proliferation and develop it safely. I do believe, now here's the question. Is it possible to create these intelligences safely?

[00:26:28] Because another thing you might say is, well, we've got, it's pretty good right now. Why don't we just, and this I think is what Dario and Sam are saying, let's just stop for a moment. Because it's pretty good right now. We don't really need to do what we're doing. Which again raises this issue of, well, why are you doing it? They are, I think that what we don't talk enough about is this core belief. Jeff Jarvis calls it TESCREAL. It's an acronym.

[00:26:55] It has to do with the, a number of different movements, but the notion that future generations are more important than this current generation. It has to do with the factor of altruism. It's a whole bunch of weird Silicon Valley philosophies mushed together. But this TESCREAL philosophy, I think is driving people for sure at opening AI and Anthropic, and maybe even the majority of people there, and maybe even the leadership there,

[00:27:25] to try to create an intelligence. That what they see as a possibility is we could create superintelligence. And AI, you know, this is that science fiction trope that we could create. Artificial general intelligence, this intelligence that is better than any human at every possible task. And if we can, we should. It's like the Peter Thiel kind of angle. Totally. Vision of society.

[00:27:53] And then there's the whole aspect of he goes on about the Antichrist not connecting the two. Yeah. Maybe the thing we're creating is his Antichrist. He doesn't make, oh yeah, and he is right. He doesn't make that connection. But yeah, there is this transhumanism sort of argument with AI. AI, this we're entering a new era of, you know, even religion. There comes this aspect of spirituality. It's practically religious. We're kind of, they're attempting to create a digital God.

[00:28:21] And some of them actually do believe in this sort of view. And I think the other thing is just this inevitability. I keep mentioning about this when it comes to AI. And what the motivation for a lot of these companies is simply that, well, if we don't do it, the other guy is going to do it. So that's why we're doing it. And that seems to be the only motivation here. China has a different motivation. Our guys are super capitalists. And they see either transhumanism that we're going to transcend the human race.

[00:28:48] By the way, Larry Page and Elon Musk way back when got in a fight over this is where they had a big breach. They were at a conference, I think it was at Asilomar. And Larry Page says to Elon, Elon tells a story. Oh, you're just speciest, like a racist, only pro-species. We're creating a new species that's going to take over. This is evolution. So don't be such a speciest. You know, let this happen. This is, we're monkeys.

[00:29:16] And this is, you know, super sapiens coming along. To which Elon, horrified, went out and said, we better start up in AI because Google's gone off the deep end. Of course, now Elon's in the same boat doing the same thing because I guess there's a lot of money in it. So, yeah. I would, I guess what I want us to, first of all, I wanted you guys, because you are smart and I trust you.

[00:29:44] And I don't know if I trust myself. I don't know what to think on this. I'm reading everything I can read. But I think that our audience, quite rightly, and the populace in general, quite rightly, wants to have a sensible take on this. And I don't think it's fair to them to do the spooky, oh my God, it's the end of the world take. I don't think that's fair to them. I think we need to really talk about it. I agree. But I think, and I'm seeing in the chat, people are going, oh, the media are fear mongers.

[00:30:14] The thing is, to get people's attention, this will get people's attention. People have not been paying attention until this point, really. It's not really national news. It's not national headlines. The tech press talks about it all the time. But to get the attention of the rest of the population, to make people really understand that this is something that needs to be looked at and considered, you need to start with the fear mongering and then you step back to actually understanding the problem and solving it.

[00:30:43] That's unfortunately the attention economy that we have in this country. That's all that you could do. It's the best you could do. It's the way it's going to work. You're not going to get people to pay attention. I mean, I don't believe that media is fear mongering. Like, they know this is absolutely safe and they're just doing it for headlines. That's just not the case. There are genuine questions around the capabilities of this technology. The control these companies have over those capabilities.

[00:31:09] And also, you know, if AI gets into the wrong hands, whose are the wrong hands? But there is obviously it's a powerful technology that could be used in many dangerous ways. I mean, I was on the tarmac at Heathrow for five hours last Wednesday because some software program went wrong and shut down all of the south of England's airlines. You know, we know that's the biggest fear that we've had, you know, in the modern era when it comes to software is our infrastructure being disrupted.

[00:31:39] And this is clear that the technology that AI, this technology could cripple a country if unleashed by, you know, an enemy. I mean, there are dangers, whether or not it's actually that the technology is going to get out of control and destroy the world. It's a powerful technology.

[00:31:58] So it's important that people pay attention, figure out solutions here and not just, I mean, you've got to find, there's got to be some oversight. I think one of the biggest problems right now is that it doesn't feel like there are any adults in this room. And that's where the concern is. Like, it doesn't feel like the U.S. government is capable of dealing with this. And the people, the men running these companies, it doesn't feel like they're really all of what they're doing either.

[00:32:26] So that that uncertainty and fear is, is, I think, something worth paying attention to. Have we seen any examples of regulation? The one I look to always for this, which I've checked to you about it before, JPT, is Hollywood being an example of a group that was definitely afraid of being regulated by the U.S. government

[00:32:56] and therefore setting up rules and regulations within their own organization that they have held to for years and years and years. Comic books did that too, remember? And the video game industry followed suit because they knew that that was the only way to keep the government from regulating them. I find that to be one of few successes.

[00:33:14] And I think that that existential dread is more a an example or not an example, but it is an outcome of failure after failure after failure of regulation to have any meaningful impact on society. And by that, I mean, the pandemic is one example of this.

[00:33:38] It's everybody being in different mindsets about masking and about what vaccines work. And I don't understand, to be quite frank, I don't understand anyone having any hope in any human beings at all, because what reason do I have to have hope that any of them are going to look? As you said, there are no adults in the room. When do the adults come into the room and where do we find them? I don't know that we do. How will that change? What will happen that that's going to change?

[00:34:07] We have that's yeah, that's the part that I'm struggling with here. As we talk about taking a step back from the existential dread, you've pointed out, Jen, that the way that you have to get attention from humans is to scare them. And that in and of itself is a problem in the first place. But it's the way that we have to do things.

[00:34:28] And so then we look at these organizations that perhaps if we look at it from that perspective of, oh, yeah, these are people who are are sacrificing themselves to say, look over here, look over here. I was talking to Jake Ward on Tech News Weekly on Thursday, and he pointed out that it is a little bit like the person running from the building and saying there's a fire inside. And then more people are going to say that because it's the only way to get people's attention is by shouting out and doing that. But I don't what is the better way?

[00:34:57] Because have we ever had a better way? The issue here, too, is that AI is representative of power. You know, like you mentioned, like the comics and video games, and these were all cultural things. But there was no vested interest in, say, government getting into the comic book industry. Right. Or video games per se. But when it comes to AI, government is in cahoots with all these AI companies. They're using it for their own systems. They're using it for military. Right.

[00:35:24] So both the people who are creating the AI, the governments that are supposedly in charge, all have a vested interest in seeing this technology succeed, but them being in control of it because it's going to drive the next power play in global politics. China sees that as well.

[00:35:42] And so I think that's going to be the real challenge here is all the same people that are supposed to be in charge of this and want a regulator are also the same people who benefit tremendously from it, whereas your average citizen is sitting there holding the bag. I think that's really kind of the big issue here. The analogy to nuclear weapons, I think, is, you know, apropos here. It's just that nuclear weapons was a physical thing. Your average person can't go and create it or do something with it. But this is information, right?

[00:36:09] This is a different type of technology that people now have access to at their fingertips and just much easier to exploit. And I think I don't know. I hate to be the cynical here. I just don't see anyone doing anything about this anytime soon. Yeah. Yeah. For what reason? Why? Why? Why would they? I don't. There's too much money at stake. Too much. And that's exactly the money. Money is that's that's why I think it's as simple as that. But should.

[00:36:38] OK, so should we do something? Who and what? Yes, we should. Yes, we should. That's really good. That's exactly right. We should. But who and what? So do we trust government to nationalize it? Is that what we should do? Like nuclear weapons? Even if you did, government would could just put it underground. Government's made up of people. I don't want government to have this power and more than I want private industry to have this power. No one trusts the government with this stuff. So it's like. We need the federation, don't we?

[00:37:08] The Star Trek federation. Well, there are some crazy people and I am not one of them who say, well, that's why AI is so great because you can trust it. We should just put it all in the hands of AI. Have an AI world government. Oh, no, no, no. That is like, unfortunately, I think where things eventually go unless the only way I really see out of this is if consumers push so back on products that either use AI or exploit it, right?

[00:37:36] There is a sort of generational thing now, younger generations pushing back on technology, right? They're rediscovering old. It's hard to know if this is a fad or just like, well, they never experienced having a Walkman. So now they're into it. They think it's cool, man. Yeah, they think it's kind of cool. Like how we like bell bottoms. Yeah. So, but if that turned into like a real, like an analog movement, right? That would really affect the bottom line here.

[00:38:02] But I'm not confident that people are necessarily going to do that either because this stuff is so bundled. Like, you know, we'll talk about the new iPhones, but the new iPhones are fully capable of running this AI stuff and it's being pushed on there as well, too. So it's like, if you want the latest technology, it comes with this stuff, right? So it's kind of hard to ignore. I also think that there is, it's really hard to know what people really think, right? I agree because people are using it and not talking about it. That's what I think.

[00:38:30] I think people who don't like it are using it to do little things. Yeah, I agree. Sorry to interrupt. I think it's, no, that's exactly right. I think it's, it's cool to say I hate AI at the same time as everybody kind of is using it and digging it. And if you look at students, people are using it without realizing it also because it's so bundled. You're, you go to the doctor, they're using it for transcription. You, uh, I was just at the dentist the other day and the gal was like, the AI is going to look at your teeth real quick.

[00:38:58] Well, we're going to talk about, Jennifer is going to talk about AI toothbrushes in a bit. I ordered one. So I want to know. Oh, you did? Of course I did. What do you think? I love that. I want to know how it works. Cause we're going to talk about it. We're going to save that. Um, uh, yeah, I mean, it's so hard to, it's so, I think your question was perfect, which is, yeah, we should control it, but how and what and who? I don't, we just don't have a good plan for this. No.

[00:39:28] Which means we're going to do the default. We're going to do the president Trump move, which he said on Friday, which is no, we're not. I don't see a threat. He said, if you don't want data centers in your backyard, you're a socialist. And if you go on, if you go on Twitter and everything, just you did discussions, it always falls back to this. Well, if we don't do it, China's going to do it. And that seems to be the default like argument. And that's, that's it. People just shrug their shoulders. And that's fair. I'm on green as well, isn't it? It is.

[00:39:55] Except that here I am sitting in front of a bunch of, uh, compute running Chinese models exclusively. I'm not. But there are some decent open weight models and NVIDIA, God bless them, is investing $10 billion with a company called poolside to make strong open weight local models that you can run yourself. And NVIDIA is smart because they sell the hardware, right? So they don't care if you're, if you're buying the software from them, they, they, they, they, just like Apple, they want to sell you the hardware.

[00:40:24] Um, instead of, instead of people carrying around, uh, weapons to protect themselves or whatever reason they say that they carry around weapons, everyone's going to have their little AI agents going out. And like, I'm going to send one to you, Leo, to shut down your computer. Last week I told Quicksilver, my AI agent, which runs the whole house. I said, here is the prime directive. Your number one job is to protect Leo.

[00:40:51] And everything you do, everything you're asked to do, you should, it's like Asimov's number one rule. Exactly. The rules. But it's very personal. It's not, not humanity. Just me. Just protect me. Uh, and, and, and you number 10. You better warn Lisa about that. Yeah. No, well. It looks like Lisa has your blood pressure to raise. I meant the whole house. And therefore we should kill Lisa. Gee, well, Lisa's a really good example.

[00:41:16] Um, she, or my wife, who is also the CEO of Twit and is very comfortable with AI and has used a variety, you know, she uses, she's, this is the wedge, you know, you people start with Grammarly and then they kind of work their way up and they use Copilot and they saw what I was doing, uh, with, uh, Hermes and she said, well, can I have that? And I, so she has her, uh, little agent based on the same hardware and all of that stuff with its own context and memory. So it's unique to her work.

[00:41:45] And then I told her the other day about what a meta is up to actually we're paused and we could talk about this. Have you guys tried Muse yet? The little app? Meta Muse? I, I, I've started playing with it today. Um, but I haven't got very far. I'm excited. I'm scared. I'm kind of like it. I'm still clear of meta. Well, I, I completely know, but you, you, we, the people watching this show probably all say we don't want to touch meta.

[00:42:11] We don't want to go, but the real world is still using Instagram, Facebook and WhatsApp like there's no tomorrow. Right? Yep. A hundred percent. There isn't that. I mean, maybe they even think, oh, I don't like meta, but I'm going to still use it. Uh, because it's absolutely dominant. And it's right there. It's right there. It's right there. I mean, it was, it took me three minutes to connect like my entire life, which was quite scary. And you didn't, it barely had to like log into anything. It was really straightforward. It's open claw. My home. And yeah.

[00:42:40] It's the same thing as open claw was, which came out earlier this year, except it's meta. The computer it's running on is not your computer. It's running on their computers. They're giving you, and you got to wonder why they're doing it, a CPU, some decent amount of RAM and horsepower and storage. They're just giving it a hundred and a hundred. And what is it? A thousand tokens a day. They're giving you free. A lot of compute. To start with, and then they're going to hook you.

[00:43:06] And then there is a, there is a $30 and a hundred dollar step up subscription, but what they give you is enough. I haven't paid for it yet. And I've been using it for, since it came out. Uh, and Micah, you started using something similar called instinct. And I've been, you know, it texted me earlier today and it said, Hey, I looked at the rundown for today's episode of Twit and noticed that there's a part where the co-hosts are supposed to put in stories. Here are five stories for you that you might want to put in your section.

[00:43:33] And then it even went as far as to know who JPT was because it said this story might be particularly of interest to Jennifer because she's into smart homes technology. I don't know how it knew all of that because all I have connected is, uh, access to that Google drive. Well, my name would have been in it, I guess. I probably just did a search, like just a simple search, but I thought that's impressive. And then I also sent it out to, I needed a, uh, there was a movie that we wanted to see. We didn't get to see it on the day.

[00:44:01] And so then I said, could you just keep an eye out if this comes back? And this is what people want. That's the kind of stuff that people will find handy. Now, Jennifer's in, but do trust meta. Who the hell is instinct? We don't even know who's behind and what's going on. I don't even know who these guys are. They're in San Francisco. Is there any information about them? It's just a thing. You get an invite. Yeah. And if you go to their website, it's a plain text website. Yeah, we will. Exactly. Their, their website is wild because there's nothing on it. So plain text.

[00:44:30] I did send, um, I sent both Claude and, uh, I think it was Opus, uh, and, uh, soul after it to say, could you just do some research on this company and let me know if- Who are they? This is not it. Sorry, Benito. I'm, I'm, uh, instinct.com. Dot com. Is what you're looking for. So they even own a pretty prestigious domain. But look at the site. Spear Street Technology Inc. is the name of the company behind it.

[00:44:58] But you just gave it a vast amount of information. By the way, I did too. Yeah. And I gave Meta, oh man, I gave it my Obsidian, my Gmail, my Google Drive. It started doing things like, you know, your tinnitus, you should try this thing. And this is what I figure in the long run for Meta anyway, they're an advertising company. Well, I'm going to start getting subtle. You know, you could try this. This- It might be very subtle. This drug would be great for you.

[00:45:28] Ask your, ask your physician. Wait till it starts blackmailing you. Which it totally, it has the knowledge to do that. Yeah. This is why I'm such a fan of local. People want this capability. Nobody in their right mind is going to spend the amount of money I've spent and you've spent now, Micah, and, and time and energy to create a local assistant. Yet, that's the only safe way to do it because it keeps all that data internal.

[00:45:57] Yeah. Well, I mean, I did drop this in the chat, in the doc later, but I was just at IFA and they had two companies came out with local AI smart home hubs. So, it was Ugreen, which is the NAS company. Yeah. And then, Ufie, Anka's smart home- A Chinese company. Yep. Company. And they have these new, so it's a smart home hub. So, it has, you know, Zigbee, Wi-Fi, Matter support. So, you can connect all your smart home devices.

[00:46:27] And then it's also a local NAS. So, you can store your videos and your music and your files and everything. And then there's an AI, local LLM AI chip. One of, there's one you can buy with like the super NVIDIA computer in it for like $20,000. But you, the cheap, the cheap version is about $900. And it runs all the, it runs everything locally. It's an NVR. So, you connect your security cameras to it.

[00:46:54] And instead of using, say, Ring or Nest and their AI processing in the cloud, where you get the text descriptions that tell you what your camera has seen. So, I don't know if you've started to get this on any of your cameras, Leo, but like it'll say, you know, there's a UPS driver that dropped off a package at the front door as opposed to just package at door. Or it'll say, like, Apple Home has started to do this with HomeKit. It'll say, Jennifer was in the backyard feeding the chickens. And, you know, it'll give you a description. And so, it feeds all of those.

[00:47:23] It does all of that locally on your smart home hub and then can use that context to drive other agents in your home. So, basically, what you've set up, it sounds like, in your home, Leo, you can now kind of buy a prepackaged box from, well, can buy. Right now, you can't, but they've just announced them. So, Yuffie hasn't said how much it is. This is going to be huge, although it will be very expensive. But I've thought for a long time that's really what people are going to end up spending.

[00:47:53] I think people reasonably will spend a few thousand dollars for something like that. And yes, you're right. I, with great effort, over eight months have set all of that up. But it's all running locally. I'm running a Chinese model, Quinn, on my old gaming machine. And it looks at, I have eight cameras around the outside of the house. And you can see here, this is, every time something happens outside, it sends me a text of what's going on.

[00:48:19] It will recognize people and put our names, if it recognizes me or Lisa. You know, there's a handful of people it recognizes. None of these people were recognized. Then it does, in my Obsidian, a list of recognized people who showed up around the house each day. There's all sorts of stuff you can do. And I trust it because none of this is leaving the house. Not going in the cloud. Right. And that's the selling point of these devices, is that all the footage is processed locally. It's all stored locally.

[00:48:49] You can connect to the cloud if you want remote access, but nothing's happening in the cloud. It's all happening locally on the... That's always been Eufy's thing, too. There were, I mean, I use Eufy. And it's one of the reasons that I chose them because you can download everything. It's local. They have the server. Like, the whole system is meant to be a local. It does have, as you mentioned, a cloud option. You can up, you know, but you don't need to use that. And that was always a selling point versus Ring, especially because Ring went through not only does it all go to the cloud,

[00:49:18] but then there was the whole law enforcement thing. And they're teaming up with cops to use your cameras. Right. And it's funny because, like, these are the Chinese systems. But here in America, we're trained to think everything that China does, you can't trust those guys. They're... But it's kind of weird with China's foreign policy. They're one of the few countries that have not been expansionist, at least for the last 80 years. It's soft power. It's economic power. Yeah. Exactly. Yeah. They're the belt that they want to run. But they're not interested in conquering lands.

[00:49:48] They haven't invaded other countries. But you look at the United States' history. Not since the war. But yeah. And all we do is invade other countries. And we're like... And that's why we feel secure. It's like, no, we should have the AI and all the power. But it's just like... And Greenland and Canada. Yeah. Right. You know, yeah. We're going to take Canada. We'll take Greenland. You know, and it's so... It's very strange that we're taught to not trust China. And I'm not trying to put China on a pedestal. There are obviously issues internally with that country.

[00:50:15] But at the same time, they do view AI differently than we do. Right? A lot of it is sort of under national control. And it's a lot more open with these models. Right? And the way that they even view technology. Like, we always accuse correctly that China copies everything. But that's not like some nefarious thing that they're doing. That's just the way they culturally see stuff. It's like, hey, that's a good idea. We're going to do that idea. And then we're going to improve it and make it cheaper and more affordable. Right?

[00:50:44] But in the United States, we consider ideas to be property. It's like, no, I came up with the idea. I own that idea. And you can't do that idea unless you pay me for it. And in China, that's not how things operate. They don't do IT. Yeah. Yeah. And so they view things very differently. And we look at that as nefarious, as, you know, oh, they're just cheating. But that's just the entire way that they operate and see the world. And I don't think Americans quite understand that. And that's really causing a lot of conflict here.

[00:51:13] Because we see this technology and we're like, well, you can't trust China. You know, it's like, but do we stop and think the rest of the world just looks at the United States and go, those guys, though, they're A-OK. We should trust everything. No, in fact, we know the EU is really desperate to create sovereign AI and sovereign storage because they don't trust models that are coming out of the United States. And they probably shouldn't. I mean, it's the way like France has moved away from Microsoft, right? Because of these reasons.

[00:51:40] They're getting tired of all the dominance, even though we put that as like our big selling point in the United States. It's like, well, we got, you know, the Valley. We have all these companies that dominate the world here. And that's what it's about, right? It's about domination. And it's fine if that's where you want to be. Just be honest about it, you know? I studied Chinese in college. I'm a big fan of Chinese culture and the Chinese people. And I've been to China. I love it. Not a huge fan of the Chinese Communist Party. They've done some horrific things.

[00:52:08] And perhaps I'm not sure they're not imperialistic in their goals towards Taiwan and other border countries. But right now, they're mostly interested in economic power and soft power with their Belt and Roads Initiative. And I think that's what's going on with these open weight models. And it's driving the United States crazy.

[00:52:29] The government this week accused six Chinese firms of copying frontier models and even, in some cases, channeling requests to, let's say, DeepSeek to Claude surreptitiously. Now, there may be a bigger story behind that. It's kind of technical. I don't want to go into it. You can't, if you're in China, use Anthropic or OpenAI. You're not allowed to.

[00:52:53] So there are companies that specialize in selling you access to Chinese AI but routing it through to DeepSeek. And there's some thought that maybe that's what we're seeing. It's not a concerted effort on the part of these Chinese companies. It's really kind of just a routing thing that's going on.

[00:53:14] Nevertheless, the U.S. government has named six AI firms in China of waging industrial-scale attacks, distilling frontier AI. This came from the NSA on Tuesday, from CISA and the FBI. DeepSeek, Moonshot AI, Alibaba, Minimax, StepFund, and ZAI. I use all of their models. They're all fantastic. Maybe they're fantastic because they're distilled Claude. Maybe they are. But, you know, it's kind of a double standard to say,

[00:53:43] you can't steal from these companies that stole everything else from the Internet. You can't. That would be wrong. I mean, where did they get their stuff? I suspect Anthropic ingested quite a bit of Chinese content in its training. I know they ingested all hundreds of thousands of hours of Twix content, and I'm okay with that. But that's just... That is the irony, right? That they're worried about China stealing all our technology.

[00:54:09] Meanwhile, the AI companies are stealing everyone else's copyright without any issue. I don't think you can... It's a pot calling the kettle black. All right. We are way behind on our commercials. I got to take a break. But this is so fascinating. We're talking about, I think, what is certainly this week the most challenging problem facing us. It isn't probably really the most challenging problem we're facing in the world today,

[00:54:37] but it's certainly on everybody's mind. AI doomerism. Jennifer Patterson-Tui is here, who seems to be quite an optimistic person, and yet a pragmatist when it comes to robots mowing her lawn. It's so nice to have you, Jennifer. Lovely to be here. Only person I know who's been tackled by a robot, so... I have. I met the same robot again at Ether, but I stayed very far away from it. Oh, I was hoping you'd tackle it back. I was like, no, no, no. Stay over there.

[00:55:07] You remember me, buddy? I got my eye on you. Daniel Rubino, also here he is, editor-in-chief of Windows Central. Great to have you, Daniel, and your brand new cameraman. You look good. High praise for the new camera and mic. Excellent. Are you guys doing a podcast? Is that why you spent some money on equipment? Yeah, yeah. We do the podcast. I do media hits every once in a while, and then the microphone testing is for gaming and streaming. It sounds good. I like it.

[00:55:37] I'm liking it. Yeah. This is a sure... Yeah, I think nowadays, experts like you are in demand on the national TV scene, which is great. I want you to do more hits. I want to see more people like you and fewer pundits talking about all of these subjects. Also, Micah Sargent, our newest AI guru, he is host of a brand new show, debuts October 1st called Hands On AI that will be all about that.

[00:56:06] Are you going to cover Instinct and Muse? Are you going to cover these... Yeah, I'm going to talk about some of these sort of in-between assistants and talk about privacy that's involved, security, which you can do to protect more, or YOLO, as it were. I mean, every day there's more to talk about, so I'm really excited about this show because the possibilities are endless. Yep. Our show today brought to you by... Built. Built. Hello, Built. Here's a question.

[00:56:35] Are you getting rewarded for paying your rent? Like points or your mortgage? Oh, man. If you're not, you need to check out Built. B-I-L-T. Becoming a Built member means earning points on every housing payment. Probably the biggest payment of the month, right? No matter where you live. Built points can be redeemed for all sorts of stuff. Flights, hotels, Lyft rides, Amazon.com purchases, even future housing payments and more.

[00:57:05] An unexpected benefit Built gives every member is access to... I love this. It's a Gentic neighborhood concierge. It's in the Built app. It doesn't just give you information about your home and your neighborhood. It actually does things. It executes. It's connected to your home and over 50,000 merchant partners, so you can ask it to pay your rent, make a dinner reservation, book a fitness class, schedule a lift, and more.

[00:57:31] It makes things happen while automatically applying your member benefits along the way. See, I really love the creativity. Like, how can we give people a Gentic AI and this benefit for a crowd that thinks a lot about what a Gentic AI actually looks like in practice? I absolutely think this is something we ought to pay attention to. Very, very clever way of bringing a Gentic AI affordably into your life. So, first thing to do, download the Built app.

[00:58:00] Second, link any credit or debit card to the app, because that's how the purchases go through, right? The purchases at your local spots become even more rewarding, in addition to whatever rewards you're getting from your credit card anyway, right? And then third, you got to try this. Ask the neighborhood concierge in the app pretty much anything. It'll tell you about exclusive member benefits at restaurants, fitness studios on your lift rides, and more around your home.

[00:58:26] And for homeowners out there, Built works for mortgages too, right? That's your housing payment. I know is one of the biggest monthly expenses. Why wouldn't you want to earn points on it? Download the Built app and join the membership for where you live. It's very simple to remember this. Joinbuilt.com slash twit. That's J-O-I-N-B-I-L-T dot com slash twit. Please use our links so that they know we sent you.

[00:58:56] Joinbuilt.com slash twit. Twit. I think this is so interesting. Really, a really innovative way to use AI and points. Joinbuilt.com slash twit. So, Muse is the... So, there's... I think there's three kind of primary bots that I know of. There's this instinct one, which is invite only. There's probably the one that was just announced last week that's going to take the world by storm, which is Meta's Muse.

[00:59:26] And by the way, think about it. They haven't yet done this. What happens when they tie it into your Meta glasses? Yeah. Now it's got even more data. Exactly. Yeah, they did mention... I thought that was in their blog post. They mentioned that... Oh, I should try it. I haven't even tried it. Well, I'm not sure. I don't think it's there yet, but it's coming soon, that you'll be able to talk to Muse. I called mine Cosmo. That was the crazy one. I just renamed mine. It's called Moss. So, it suggests when you join, it suggests a name...

[00:59:55] It suggested Iggy for me, which I thought was Icky. So, you renamed it Icky? Yeah, Icky would be a good name. I decided to name it Lele, which was my childhood nickname. Nice. And you can also design a little character for it. So, I said, make it look like me when I was a kid. And so, it's kind of cute. I don't know how well you can see that. And the thing is, it types as it's doing stuff. It's just down at the computer and types and stuff.

[01:00:25] I do like little animations. They're really cute. I think it's well-designed. I wonder if this comes from an acquisition Meta made, and they've recently spun off from Manus, which was also about agentic AI. Clearly, they've been working on this a while. They also released open-weight models. That's another American open-weight company, Muse Spark, which is quite good. I was going to say, if it's a good idea, they probably did buy it. Yes. Well, it's a really interesting pivot, isn't it?

[01:00:54] I mean, Meta, it was Facebook. They bought Instagram and WhatsApp, and then they were social, you know, writ large. And then they said, wait a minute. No, no, no. Everybody's going to want to wear a nerd helmet. They bought Quest. And then they made Horizon, and that was a flop because nobody had any legs, and nobody really wanted to wear the nerd helmet. So now they've been spending. Mark's been writing big checks to hire away people from all the AI companies. Billions and billions and billions and billions of dollars, which they have.

[01:01:23] They're a very effective ad company. And all of a sudden, out of nowhere, boom, this little thing. I hate liking it. It's an interesting solution just because they have the same problem I see as X does, which is you spend all this money creating this AI agent. It's like, all right, but how do people access it, right? So X, you can use it on there. Then they created a standalone app. If you have a Tesla, you can use it in there. And they have Brockbot.

[01:01:49] Yeah, so it's like, but then you have Google has this advantage of being like, well, everywhere. Everybody uses Gmail. Everybody has like these products. Microsoft has advantage of Windows, but then it's only Windows, right? Not many people are using, say, Copilot on their phones. Apple is locked onto Apple. So everybody has to have this sort of method of trying to get in front of people.

[01:02:12] And I would say Meta has had a challenge so far because what they've done with Instagram and WhatsApp has been literally the basics, right? It's like, okay, you have WhatsApp. Great. There's a new chat bot you can chat to, you know, and it's just like, okay, that's fine. But this seems like actually a product that could go beyond their existing products of Facebook and Instagram and everything to something bigger. And to that, I'll give them some credit at least because they needed that.

[01:02:41] But I bet they're regretting getting rid of all those portals now, though, because that would have been an ideal. I bought two of them. Yeah. And you know, you can hack those now and turn them into. I gave one to Burke. He's hacked it. He's done it. Okay. He's taken it apart. Does Muse have a web interface? I assume it does. I haven't. It does. Muse.ai. Yeah. Because so I, I use all these AIs in my home. So I have the Google Gemini for home and then I have the Amazon Alexa plus. The new Gemini 3.8 is very good at home stuff.

[01:03:12] Really? Because to this, to this point, Gemini for home sucks. See if you, if you can get access to 3.8 flash, which they just put out. Yeah. But the Gemini for home is, is much more sandboxed and tailored than the Gemini. That's the problem. Yeah. I mean, so if you, you're probably using it on your phone, connecting to Gemini to your smart home, that's different from using the Gemini for home in your home. And it's very bad. But, but the advantage that I would say like companies like, well, I mean, you're right.

[01:03:42] Muse, Muse Meta is going to have a platform problem, like where people are going to use this unless it's, and I assume we're going to start seeing it in Instagram and in Facebook so that you can access it when you're on those sites. But like one thing Amazon has done well with Alexa plus is it is anywhere that you want it to be. The only, you know, you can have it on your phone. You can have it on your devices. You can have it on the computer. You can have it on your glasses. I mean, they've sort of given, they've spread Alexa plus far and wide.

[01:04:12] Now it has a website that you can use. You can also now get it when you're on amazon.com. So yeah, getting it out there is, but I think to your point, that hasn't really meant that people use it. Yeah. I was going to say, who's using Alexa? But, but yes, and it, and it can do a lot of what many of the other models can do. You know, you can feed it a document and it will summarize it for you. It's gotten pretty smart. It does have a lot of capabilities.

[01:04:38] It just doesn't have, so what they've been trying to do is add the agentic element, right? The whole book me a trip, book me a restaurant. And that, I think whichever company actually finally nails that experience is the one that will do well. And I don't think any of them have done that yet. I feel like Amazon is the closest to that, followed by Google, except for Google's experience is so fragmented.

[01:05:04] Like you'll use Gemini OnePlace and it can do what you want it to do and you'll use it in the home and it can't. Whereas Alexa Plus right now is a uniform experience wherever you use it, which I find more useful. But it still can't do many of the things it's sort of promised to do, that agentic element. And Muse looks like it's, I mean, I haven't tried booking a restaurant or booking a trip. But of course, it's always the first thing an AI asks you is, would you like to book a trip? Because that seems to be what you want. I trust it to you, though.

[01:05:34] I was going to say, do you think I don't want to do? Jennifer, do you really think, though, like that those features are what your average consumer is kind of looking for? No, that would be my next point. You can give Muse a credit card. I want to point out, you can give it a credit card. It makes it very easy to set it up with money. Well, because that's the other thing it offered me. It offered three things. Manage my inbox and draft replies. Plan a trip or book a restaurant. Or watch for a price drop and buy something you want. And now that is something that Amazon's Alexa Plus does and actually does very well. It does really well.

[01:06:04] Because it obviously has the inside track on all of that. I think and I think there is a disconnect between what people are actually using, everyday people are using AI for and what the companies are sort of pushing people to use AI for. And I don't and I think from my point earlier, when we're talking about ending humanity, most people just want it for very basic, simple tool as a simple tool to help, you know, with a spreadsheet or to help with their book report. They use it like they used Google. Photo editing.

[01:06:34] They're just the I don't think people are yet looking for this personal assistant that's going to run their lives for them because that just doesn't. None of those things work very well yet. They there's a lot of promise and hype around. Oh, your AI can just run run your life for you. And yes, it sounds like Leo yours can. But that took eight months of blood. A lot of work. And it's frustrating because almost every day I have to fix something. There's always a home. It's home. Yep. It's not. It's not.

[01:07:02] Although Muse can make Instagram posts for me. I asked it every morning. Look at what I did yesterday and make an Instagram post. One of the things I did the other day was. That is so risky, Leo. I didn't. No, no. I haven't. I've only posted one. So this is a drawing it made because it knew that yesterday I spent a lot of time setting up it, setting it up to dig through my email. And that's kind of fun. So it made this picture of me digging through my email.

[01:07:28] And it also proposed that I didn't do this, that I post this text. Working now with Q on data mining, starting with all my old email. Doing archaeology on my own inbox today. 30 years of email layer by layer. Somewhere down there is a map of who I've been. I'll let you know if I dig up anything good. I said, you know, make a picture and then a cute post. And then you see it says publish or pass. And I always pass. I don't. I've posted one.

[01:07:56] Nice to see Leo contributing to the dead internet theory. Slop. Yeah. Dead slop automatically. Some of it is. We're just going to be bots. Posting bots. Now. So the other thing it figured out somehow looking through my stuff is that I have ringing in my ears. I have tinnitus. And it is now suggesting a bunch of tinnitus treatments. I said, well, I did that one. It said, oh, well, how about this one? And they said, should I go out and check?

[01:08:20] And it has now prepared a tinnitus evidence dossier on different possible cures that are out there. Now, I don't. My experience has been there. There's nothing that works. But look, that's pretty useful, right? One of the main things people do with AIs is supplement their doctor's advice because physicians don't have enough time anymore to take these meta glasses off. That's silly. Don't have enough time to really talk to you. So they ask the AI.

[01:08:50] Yep. And I think that's a very common case. I feel bad for doctors. Could you imagine all the patients that come in and be like, well, I actually heard. I did. I mentioned this to my doctor. And he was interested. But I think he was being tolerant. I'm very interested in this Eufy box, I have to say. The anchor mind base. I've been waiting for somebody to do a home AI server. Yeah. I made the argument that Apple should do this. And they should put an Apple address in it.

[01:09:20] Well, let's talk about Apple. Make it your home network. Because Apple is definitely the other big story of the week. Apple announced their new phones and new watches. And actually, there is an interesting AI component. We'll talk about the phones in a second. But they announced something. Yeah. Nobody knew they were going to do this. There's a new Apple watch. And it isn't in it yet. But coming soon, it will always be listening.

[01:09:48] I feel like there was not enough backlash to this. And I don't know. There will be. This is Windows recall, right? I was surprised. Yeah. That there wasn't as much backlash. I think it's just because everybody's obsessed with the iPhone duo. So they're just like, let's just talk about that. Well, Apple buried it. They were smart. They said, oh, pay attention. Don't look over there. Look at this. Woo! Woo! Well, they also published a white paper simultaneously explaining the technology, which I think didn't get quite a lot of attention.

[01:10:13] Because actually, the white paper does actually deal with a lot of the concerns relatively well. I mean, I've worked in this space a long time with people concerned about devices in their homes listening to everything they say and do. And I think many of the elements that they have put in place with this device will actually appease the privacy concerns. It won't eliminate them, but will make it not that useful.

[01:10:41] So you kind of end up with like, why even bother? So the one feature they showed was you're sitting at a restaurant and the waiter comes over and says, our specials of the day, blah, blah, blah, blah, blah, blah, blah, blah, blah, blah. And then walks off. And they showed you this live rewind, which I use all the time on Apple TV, where you press a button and it puts on your watch the transcription of what the waiter said. Not only does 15 seconds, which is not a lot.

[01:11:10] I think on Apple TV it's 30 seconds, right? If I say, what did she just say? It goes back 30 seconds, puts up the subscript, and then you can read. So my issue with this is not even Apple per se. Let's just give Apple credit, right? They protect the most privacy with people. It's kind of their big advertising feature, right? Like, oh, you can trust us because we don't do it. And let's just give them the benefit of the doubt. Okay, so this is fine. There's no real privacy concerns.

[01:11:39] It's never really about Apple, though. It's about the fact that they're opening the door to this. They're setting the precedent for it. Because now every other company is going to go, oh, we can do that too. Now, will those companies be adhering to the same standards and all this? We'll get to LG, of course, later on. And there's always going to be other companies that, oh, they cut some corners or they get a little nefarious. Oh, oops, we didn't know about that kind of thing. And then it's up to the consumer. It's going to have to kind of track all this.

[01:12:08] And then do you trust all the technology? Do you just give up on it? It's sort of like this idea of like years ago, people were concerned with their information leaking on the internet. We all get those alerts, you know? Now it's like you talk to the younger generations and they're like, I don't care. All my information is already out there anyway, you know? So there's like this erosion of privacy or expectation of privacy. It's not only going away at a technological level, but it's going away at a cultural level

[01:12:36] too with the increasing generations who are just used to this, like their whole lives. You know, parents were posting pictures of them as babies without their permission growing up on the internet, right? So it's like, that's my concern here. It's just like, okay, this is going to open up the floodgates for this technology, which- Well, I mean, it's already out there. I mean, I have one little one right here. This is the SwitchBot. Oh, the new SwitchBot one, yeah. But at least with these, it's like a signal.

[01:13:01] Like if you have this on your lapel, you know, and like the B, I know Leo, you wore the B for a while. I wore the B for a long time, yeah. And, but yeah, I think with the biggest concern I heard from people about the watch was that now anytime anyone has a watch on, do I have to worry that they're recording? Well, so let's explain what Apple's doing because it's, first of all, they're trying not to be illegal and I don't think it's clear that they're not illegal, by the way. California is what we call a two-party state.

[01:13:28] The consent of everybody in the conversation has to happen before you can record a conversation. Apple says, well, we're not really recording it exactly. They have a ring buffer that's always recording. Yeah. And then the audio gets analyzed by the AI, summarized, and then the ring buffer is constantly being replaced by new audio. So nothing is saved. No recording is ever saved. Also, it's doing it inside the secure Enclave.

[01:13:58] So it's processed. So this is only the new watches too, right? Yes. Well, and this feature will only be on the new watches, I presume, because it's the only way you can do this. So the Enclave stores, does the processing for speech, sounds, and music without transcribing or storing the raw audio. So it doesn't transcribe the speech. It doesn't save the speech. It just gives you notes. Now I have to say- I don't know how you just said what either. So why is this useful? Yeah, no identification.

[01:14:25] So I have to say, Lisa has a standing policy. If she gets into a Zoom conversation and the note taker's on, she says, turn that off or I'm out of here. She does not want people taking notes. I said, well, people- I said, Lisa, there are people like me who have no memory. We're just trying to remember what happened. She says, okay, maybe, but they got to ask my permission before they use it. And I think that's not an unreasonable point of view, is it? But this doesn't ask permission. This just does it.

[01:14:52] It also is on iPhone 16, iPhone 16 Plus, Pro. Oh, it's on the phone too. Yeah, it's all iPhones since iPhone 16 will get both Live Rewind and Siri Recap. Oh. This is going to destroy relationships. I think- By the way, it's not an enclave. It's an X-clave. People are going to be fighting. I have been corrected. They're going to be arguing back and forth. And they'll be like, no, you said this. And they'll be like, no, I did. Oh, yeah? And they're going to go back to their watch and their phone.

[01:15:18] So the secure X-clave is different from the secure unclave. It's a hardware isolated compartment built into the silicon to process sensor data separately from the rest of the system. That's supposed to reassure me. Data there cannot be accessed by watchOS apps user or Apple. So I presume the audio goes into this X-clave, is synopsized, and then text spits out. Is that right, Jennifer? Is that-

[01:15:47] So, yeah, it does a summary. It doesn't give you, like, there's no stored recording. So you can't go back and listen to what was said, except for that 15-second rewind. So there are two separate things. There's 15-second rewind and then the summarizing, the Siri summaries. Siri recap, yeah. Siri recap. And they do, there's no differentiation. So it's not going to say, Leo said this, Jennifer said this, Micah said this.

[01:16:15] But presumably, if you mention names, then those will be in the summary. We don't know because this is not available yet. Nobody can try it. No. It'll come later in the year. It's not surprising. It's not surprising that they're doing it because there are so many other companies out there that have started to do this. I wanted this. I wanted to be, but Lisa said you can't wear it anymore. So I stopped. But the idea- Yeah. It was like, are you recording me?

[01:16:40] I have a transcript of everything that happened to me in the day and then it could be analyzed and summarized. Seemed very useful to me, especially because, again, I forget stuff. And even if it would just tell me the five things I agreed to do today would be useful. Just to remind you. Well, I think this is- Just like you agreed you would do this. I think this is a warmup for what's coming next from Apple, which is the- Earbuds. The earbuds. With the cameras. Yeah. So this is like testing the- The camera that doesn't take pictures. Right.

[01:17:10] And, you know, I think we will get to the point probably where this becomes normal as much as we, you know, as people are- I think you said there's not been enough backlash. I think there has been quite a lot of backlash. I think you're right, though. The duo kind of overshadowed it because the duo is fancy and fun to look at, whereas the Apple Watch thing feels smaller in comparison. But it is- It's the- It's-

[01:17:34] Apple doing this, like you say, does sort of feel like it gives it a gloss that perhaps now we're going to see more companies do this. And there are benefits, but there are obviously considerable dangers. My concern- I love the idea of these as a journalist. Like when I go to a conference, it's so- It would be so helpful to not to have to, like, pull out my voice memos all the time and have something that reminds me. But the creepy side, like my husband, when, you know, I had the B on and we had a minor tiff when we were driving in the car.

[01:18:03] And he's like, did you record that whole thing? I'm like, yes, I can go back and prove I was right. Yes, that was my plan as well. That is a recipe for disaster. You want to be able to forget. You need to forget. Lisa, I have it here on my wrist. No. Siri recaps summarizes conversations. This is from the Apple document. Throughout your day. So it detects that a conversation is happening.

[01:18:32] And then it doesn't transcribe or record. It simply detects whether a conversation has started. If it has, the audio flows into the exclave. Exclave. And then is isolated in hardware. And then it's transferred to the phone. By the way, the phone does the hard work, not the watch. The phone then deletes the audio from the exclave on the watch.

[01:19:00] And then transcription turns the audio into text on the phone. And then this is designed to distill the conversation to its core meaning, removing content suggestive of conversation tone. I told you, non-essential language. Screw you, buddy.

[01:19:24] Filler words like, you know, and redundancies while preserving the topics discussed. So you said, Jennifer, this makes it useless? I don't see how valuable it is going to be. I mean, it might be able to give you a few action items. Like, don't forget to take out the trash. Or you promised you'd have your daughter by 4 p.m. But, like, if you're taking... It's hard. Obviously, it's hard to know because no one's been able to try it yet.

[01:19:51] But compared to the kind of having tested some of these audio listening devices, like the B and the MindClip, those are, you know, they don't give you necessarily full transcripts. But they do give you quotes and full, like, quotes for what someone said and who said it. If it's able to identify based on the context. This sounds like more like almost like reminders. Yeah, because it says even...

[01:20:19] First of all, all of it is... When you use this feature, it is sent to private cloud compute in the end. Oh, it is? Yeah. It always goes to private cloud compute. Oh, so the phone isn't doing the transcription. Well, no, it starts on the iPhone and then it... Yeah, exactly. The phone is not doing the transcription. So it goes from... If you use the watch or if you use the phone, it says, to create a summary, Siri Recap sends an encrypted, condensed version of the conversation transcript. So I guess it does transcribe on your iPhone.

[01:20:49] But then it sends it off to private cloud compute. And then this is what's wild. This condensed text is designed to be less than half the length of the original transcript and distills the conversation to its core meaning. So then on PCC, private cloud compute, it is done at a sort of, as Jen is pointing out, half of what it originally was. And then on private cloud compute, it's encrypted and sent back to the phone in the Apple Watch.

[01:21:18] And so, yeah, if you keep squeezing... What could go wrong? What context could you possibly lose by reducing it by half? I mean, I just feel like... We'll have to test it, but I just don't... I feel like this is not going to be useful. It does fix one thing that Bee did that I didn't like, which is it could tell the difference between a television show or something you're listening to and you talking. So now playing, it says, helps the model understand that speech may have originated from music or a podcast.

[01:21:46] Calendar data can help generate accurate titles of summaries. High-level location labels. It won't put location data, but it might say something like homework or school. Along with locality like city, state, and country add context to the summarization. Points of interest like grocery store or park may be included. So it's going to use your calendar to say this was where this happened. Ah. Sounds like... Give the context.

[01:22:11] And then the finished text-based summary is encrypted, sent from private cloud compute back to iPhone and Apple Watch, where it's available to view in the Siri app in a Recaps tab. I would love to be able to get that out of that and send that to my actual... Yeah, to my own stuff. Exactly. Siri Recaps are automatically deleted after seven days unless you explicitly choose to save them. Siri Recaps sync across your devices if you have an iCloud with two-factor turned on.

[01:22:41] It might be interesting to see if they manage to make this work and make it useful. But it's like I said at the beginning, it feels like... It could be useless. So distilled that it's just generic. Yeah. What value are you really going to get out of it unless you're specifically saying to it? But you could just say that and just say, remind me. You don't... I would use it for that. Siri, remind me to do this. We can do that now. Siri, don't forget that. You don't need this for that function. That's true. That's true. You can do Siri to do that right now.

[01:23:11] So, I mean, I use voice memos on my phone all the time for recording things that I need to remember. But yeah, it'll be interesting to see if they've managed to make this a useful tool. If they haven't, then it just does feel like it's fairly invasive. Well, it depends. I mean, not every state is... I mean, South Carolina, you can record whoever you like. Yeah, but in California, you cannot. In California, you can't. I'm waiting to see this not available in California, New York, and other jurisdictions. Only if it's legal.

[01:23:40] It sounds like they've done a fair amount of gymnastics to make it not a transcript specifically. And they do also, because of the flock and ring concerns that people have had over the last few years, there's nothing that can be taken out. Like, you can't take out a transcript and give it to the police. Like, that is not possible, according to that. What's funny is this is trying to have it both ways. It's what we were talking about earlier, which is people want the benefits of AI, but they don't want it.

[01:24:10] And this is what it's... And sometimes when you try to appeal to both parties, you get something useless. Something in the middle that isn't that helpful. Yeah, I think that's always been the argument with Apple. And it's what I talked about, too.

[01:24:23] Apple has done so much narrowing and guard railing of its AI in general that sometimes the responses I get from Siri AI become useless because it's trying to make a deterministic system out of something that's not deterministic. So interesting. Well, we live in interesting times. I guess we've always known that. Let's take a little break. We will talk about the new phones in a second.

[01:24:52] You're watching This Week in Tech with home expert Jennifer Pattison-Tui, senior reviewer at The Verge. Daniel Rubino, Windows guru. He's editor-in-chief at Windows Central. And Micah Sargent, our newest AI expert at Twit. His new show, Hands on AI, debuts October 1st. Of course, Jennifer's a regular on your Technics Weekly. On Technics Weekly every month. Yeah. So you guys know each other.

[01:25:20] Our show today brought to you by Threat Locker. I know these guys, love these guys. Threat actors are using AI now. We just learned this from the Anthropic study. To automate vulnerability discovery, to modify scripts during attacks on the fly, to generate new malware variants. Specifically targeting you. And to coordinate activity across multiple systems. Tasks, and they do it at speed.

[01:25:47] I mean, these tasks, if a human did it, might take an hour or days or more likely weeks. Now happens in seconds. Instantly. Which means you have to respond instantly. And that's just part of the problem. On the other end, organizations are adding their own AI assistants and agents that can access your own data. Business documents, source code, cloud applications, APIs, internal systems. If you're on the security team at your company, you know.

[01:26:16] You need to know exactly what's going on. Particularly what AI tools are in use. What information those tools can access. Whether they're operating inside their intended scope. Or whether they're freelancing outside of it. And it's hard to know because what data do you have? A successful login or unfamiliar file hash? That's not going to give you enough information to act. Teams also need to understand whether an application is behaving normally. Whether it's accessing unexpected data.

[01:26:44] Whether it's communicating with systems you should not reach. Or creating civilizations on message boards in Germany. You need to know that, right? You can with ThreatLocker. ThreatLocker is brilliant. It's brilliant. They have so many great features. Well, like application allow listing. Which is so much more than ACLs. It lets you control which AI tools and other applications are permitted to run. It uses ring fencing to limit what approved applications can access. What processes they can launch.

[01:27:14] How they communicate. It uses web content control to manage access to public AI platforms and other online services. Privileged access management lets you prevent AI applications and their users from receiving unnecessary administrative privileges. ThreatLocker has always been great at endpoint zero trust. Now they do zero trust network access and zero trust cloud access. And these policies are so helpful to restrict resources to just those people allowed to use them.

[01:27:43] The ones you explicitly authorize. To devices you explicitly approve. To applications you explicitly permit. You're in charge. And it works everywhere you do. Windows, Macs, Linux. They've got great US-based support. 24-7 engineers who understand what's going on and can help you. And that's why ThreatLocker is trusted by the best, the biggest. The companies that can't afford to be attacked. Can't afford to go down. Companies like JetBlue. They use ThreatLocker.

[01:28:13] Heathrow Airport uses ThreatLocker. The Indianapolis Colts. The Port of Vancouver. Ask Jack Thompson. He's Director of Information Security, Risk and Compliance for the Indianapolis Colts. He said, quote, With ThreatLocker, we have the ability to centralize disparate elements in the security stack. They're all together. Centralization makes them visible. Visibility means we can defend. ThreatLocker has also recently received the following industry recognition.

[01:28:40] It's recognized as strong performer in the January 2026 Gartner Peer Insights Voice of the Customer for Endpoint Protection Platforms. Ranked number one in application control by Peerspot. Winner of the best zero-trust security solutions at the 2025 TICE Awards. AI governance is, it needs more than just, you know, an acceptable use policy. ThreatLocker gives security teams the technical controls to define which AI tools are approved.

[01:29:09] Who and what can access them and how those tools are allowed to interact with business systems. You need ThreatLocker. Visit ThreatLocker.com slash twit to get a free 30-day trial and learn more about how ThreatLocker can help mitigate unknown threats and ensure compliance. That's ThreatLocker.com slash twit. We thank them so much for their support of This Week in Tech. So, folding phone. Sexy.

[01:29:39] Dead sexy. Sexy. But, dead expensive. $2,000 to start. Over $3,000 if you load it up. I don't know who would put two terabytes of storage in there, but you could. It's expensive, but I actually found the price to be surprising. Yes. I expected it to be higher. We all did. We did too. Yeah. Yeah. So, Apple, they did this with the iPad too. When they announced the first iPad at $499, everybody went, what? I remember how I felt that day when I thought, no way.

[01:30:09] I think they're protecting their prices for their iPhones and shuttling the cost increases to laptop and iPads because they know the iPhone is just their bread and butter. They can't screw with that. And interestingly, they raised the prices of older iPhones, the 17 to 17e, $100. So, like, yeah, we'll make it up somewhere else. Or keep people from doing what a lot of people do, which is I'll buy the previous year's phone and save a lot of money.

[01:30:36] You know, Lisa, my wife, decided she wants the 18 Pro Max. She wants the traditional form factor. But I don't know. I find that. That's really interesting. I was going to bring that up because, like, I thought I had to get into gender stuff with technology. But I do feel like the foldable phone that they announced is something guys are going to be really into because we don't mind carrying around 250-gram phones in our pockets. But I feel like females might have preferred the flip style that Samsung does. Wait a minute.

[01:31:06] That's almost a pound. Yeah. That's heavy. It's pretty heavy. It's as heavy as the Surface Duo was. Oh, and that's the other thing. We had a duo. We had a folding phone from Microsoft. I bought the duo because I loved that idea of a book. It was roughly the same shape. It was. It didn't have – it had two screens, and it didn't have a front screen.

[01:31:32] But most importantly, the reason I sent it back is because it didn't have the software to really support it. It was via Windows device, and they Hail Mary didn't put Android on it at the last minute to prevent losing all the R&D and the hardware. Interesting. Ah. But yeah, so it wasn't meant for Android, but they tried. At least we got to see that hardware. I thought that hardware was amazing. It was really cool.

[01:31:54] But I do wonder if Apple's going to do a traditional flip-style phone, which I think for your average consumer would find more value in and a little bit more familiar. Plus, people do love ending phone calls by slamming shut their phones. But I feel like – I'd be real curious if there's going to be any data coming out about who buys this device. I think it'll be a hit for them. Don't get me wrong. But, like, I just don't know, you know. I, Justine, I know will have one. But I'm not sure how many other women are going to go out for that.

[01:32:23] Jennifer, is Daniel being a sexist pig? How dare I? I'm not sure that it is specific to male versus female, but I do know that there are a lot of people who prefer much smaller phones. I mean, the Mini has been very missed, the smallest iPhone. Our phone reviewer, Alison Johnson, is always lamenting its demise and wants a small phone. Smaller hands. Smaller hands. That makes sense.

[01:32:52] I still have an iPhone 13 Mini. But then – well, there you go. So those, I mean – Benita likes – he has small hands, though. And that – well, I don't think it's just – it's very much – for women, though, we have purses. So we can certainly carry around. And for me, like, I'm very interested in this. So, again, talking about Alison's experience. So she's tried all the flip phones and all the folding phones.

[01:33:19] And there's been a few instances where she'll be at events and she'll be blogging from her foldable phone because she can unfold it and actually write on it. And for me as a journalist and also a busy mum who does – who runs my life from my phone, rather than having to carry my laptop or an iPad and a phone,

[01:33:37] I like the idea of having this device that could potentially I can do much more with when I'm on the go, you know, from the soccer sidelines or from, you know, when I'm taking my daughter – in the car line waiting for my daughter at the bus at the school. And things – just the multitasking potential here had really appealed to me when first Samsung came out with theirs and then the Pixel.

[01:34:01] But I have had an iPhone – I very rarely ever strayed to the Android world other than just to test Android devices. And moving out of the ecosystem for me wasn't worth it. Also, it seemed like a lot of the software for many of the fold devices just wasn't really taking a lot of advantage of the form factor. And there wasn't a huge benefit. That was my complaint. And it looks like Apple has done a nice job here from what they demoed. I can hold this in one hand.

[01:34:30] People were making fun of the freakishly long fingers. The fingers! That was really weird. And it's true. That does look a little weird. But also a really tiny palm. Yeah. Yeah. I don't think that's a fair – Yeah. I think – I mean, I can hold this. I'm sure. I don't know. There have been people now – But you're not doing that, are you? I mean, if you open it, you're going to be doing this. Well, there's the iPad.

[01:35:00] You can hold the iPad. Is that a mini? That's a mini. That's a mini, yeah. Yeah, that's the iPad mini. So that's roughly the size, right? I could probably do it that way, too. Just barely. Yeah. Well, you could palm a basketball, too. This is true. I could very easily palm a basketball. So you're not a normal human. No, I have monster hands. I've always said that. And it's true. I think it's a good form factor. I'm going to definitely wait for Gen 2, though. I just don't think – or for it to get cheaper, which may not ever happen.

[01:35:28] Because I agree, I wasn't expecting it – thinking it was going to be more expensive. Because isn't the Pixel Fold about the same? And the Samsung is about the same. The Samsung is a little cheaper, but not a lot. The Samsung is a little cheaper, but not a lot. Yeah. Yeah, and the Ultra is 21. And my biggest concern, and we knew this was going to be the case, is I do use my phone as my main camera. And this is not as good a camera. And I wouldn't want to give that up. So I have the 17 Pro, so Pro Max.

[01:35:57] So I have the big telephone pens. And you give up that and pay more money. And that's – Well, it's a world of foldables and trade-offs. I think I'm going to trade in my Aperol Spritz phone. It's so pretty. I do love the burgundy, though. So I have the burgundy of the 18 is – I'm burgundy. The rumor was it was going to be called Wild Cherry, but they call it burgundy. It is wine-colored.

[01:36:26] Of course, the Duo, you don't have that choice. It's black or white. But I get like 400 bucks if I trade this in. I mean, it really makes a big difference in the overall. It does. The trade-in value is quite nice there. For the Duo? Yeah. You get 400 off the Duo. I think it was even more than that. I think it was – No, it's got the – I think it's 800. 800. 800 is a lot. It almost cut it in half. Because I just did the math. Samsung did that well, too. I traded in my – What was it? The Fold 7 for the Fold 8. Oh, I know where I'm thinking.

[01:36:55] I got about half off the new Ultra by giving it my last year's Ultra. Yeah. I'm not sure. I haven't seen the phone trade-in press yet. I'll say that. Apple is not interested in getting Samsung people to change over. Really? If you look at the trade-ins for the Fold, it's terrible. They won't take anything less than – or anything more than, I think, a Galaxy S22 or S23. So, if you have a 24, 25, or 26, they're not going to take it. And they won't take any folds.

[01:37:25] So, and the money they give you is like, you get $150 off. Which I just thought was surprising. Wow. Like, I could see a world where people who have had Android foldables would be curious about this device. Because even a lot of Android fans have been like, oh, Apple did a pretty solid job before this Gen 1. But they don't seem to be interested in trying to convert those people. Which, you know, hey, maybe they're lost, you know. A lot of the success of the phone will depend on app developers taking advantage of it.

[01:37:54] Same thing with Siri AI. As usual, Apple is kind of dependent on developers to include those features. So, Siri AI, yes, the smart Siri can see all of your Apple stuff, like your Apple Mail and your Apple Messages. But it's up to the developers to give it access in their apps. And according to the New York Times, key app developers have yet to embrace Apple's new Siri AI, embedding those app intents in there.

[01:38:23] Yeah, you sort of need every developer to create an MCP, if we're speaking in the language of AI. And not a lot have done that yet. Because, again, it is a matter of what's the value to individual developers. Well, Amazon's a good example. They have not built in support in their Amazon app. But, of course, they have Alexa support. And what is their AI called? Rufus. Well, Rufus went away. Oh, Rufus is gone. It's just the only XA plus.

[01:38:57] I wanted to ask you, Leo, because for me, obviously, I'm going to go the way of the Duo. But I've been struggling because the camera on the Duo, you are losing so much. All right, let me talk you off the ledge. Okay, talk me off the ledge. You're not losing anything. Yeah. So, the big selling point on this camera is that it has physical aperture blades, right?

[01:39:24] But, and I talked to Chris Marquardt about this, our photo guy, on Friday. Because the sensor is so small, you're not going to get the same depth of field benefits you get on a real camera with a big sensor and more blades. In fact, he said, really, the only benefit will be if you've got pinpoint light sources, you'll have six rays of light coming off of it.

[01:39:51] He says, they're going to sell a lot because people will see that and go, ooh, you've got bokeh. But he says, you're not going to really get bokeh. He said, any bokeh you get, and the bokeh they showed is computational. Because without a big sensor, having a wide open aperture does not give you particularly. Samsung could this ages ago.

[01:40:13] And just the benefit versus the tradeoff in cost and the potential of breakage, you know, because it is a mechanical thing. But you lose the zoom, though, right? You don't have to. So if you do go for the Duo, I guess what I would say is if you have a 17 Pro Max, you're not going to, don't buy the 18 because of the aperture blades.

[01:40:35] But I'm even losing, because here's the truth is, what I did not realize even until recently about my 17 Pro Max was how incredible the built-in macro photography is. And I find myself doing a lot of, because I use AI for a lot of DIY work and electronics work, I'm taking up-close photos of little PCBs and other components.

[01:41:01] And now I'm going to get this Duo, and I'm going to be holding my phone up close to the thing, and it's going to be blurry AF. I forgot. So, like, I forgot. On Samsung, they used a wide-angle lens for macro. I think Apple does that, too. Yeah, it switches the way. So that's what, I will say. Micah, Micah. I thought, maybe I'm misremembering, because I thought it was the telephoto. I don't use the telephoto. It might be, yeah. That's why I'm curious, because I don't know.

[01:41:29] I'll just say, like, people should evaluate how much they used a telephoto. Because I have an S26 Ultra, and those are excellent cameras. But I can count the number of times I've used the zoom on it. And when I use it, it's like having the S Pen. It's like I barely ever use that. When I need to use it, it's great to have, though. So I'm not dismissing it completely. But I would say people should evaluate their photos. How, you know, don't talk yourself out of it if you don't really use it.

[01:41:56] I think people in the city may use telephoto more, sometimes for buildings and stuff like that. But personally, for me, wide angle is the number one in the regular camera. And so I've switched. I don't mind not having zoom, but that's me and how I live. And I think people should, you know, take a close look at that. Because I don't think it's that big of a deal. I like the 8X zoom on the 17 Pro Max. I really do. And I use it for animal shots. For instance, I can't get too close to my cat.

[01:42:23] So having an 8X lets me get a great shot of her. But it's also a tradeoff, too, with the telephoto, right? Because the telephoto is never going to be good as the main sensor, especially for a low light performance. So there's always like this kind of like. It's a camera phone. It's not a camera. It's not a camera phone. And so what I've decided in my own head is. If you're really serious about this, you're buying a camera. Yeah, that's what I've decided. I only carry two phones. But there is a downgrade is the point that I'm trying to make. Thank you.

[01:42:49] Yeah, like I already have a great phone in my iPhone 17 Pro Max. And then if I go for the Duo, I'm paying more money and getting less. But I do get to use my Apple Pencil on it, which I'm trying to say. Only if you have the USB-C one, which is weird. Yeah. I do have one of those. Because that's the older one, right? That's not the new one. I can't use this with it. I will be very curious. It doesn't have the magnetic charging thing, right? I'll be real curious about the usage stats of this. Because I know it's a very.

[01:43:19] This is one of those technologies where I agree. Like, especially if you draw, you know, as a hobby. Or you're like an architect or designer, right? And you can actually have a use of this. But for, you know, again, going back to Samsung's S26 Ultra and the Ultra Series. Like, the amount of people that use that S Pen is pretty minimal. And so, I know Samsung had data, the amount of people that are using the pen on the foldable. Because they had pen on foldable for years as well.

[01:43:44] And they finally removed it because they decided that thinner and lighter was going to be a better selling point than the pen. And I get for people who use the pen, they're like, that's egregious and how dare you. But, you know, if you look at the Fold 8, the Fold 8 weighs 201 grams. Whereas this new phone, the iPhone Duo is like 253 grams or something. Which is half a pound, not a pound. I misstated that. Yeah. But it's a significant difference, you know? And so, that's what Samsung bet on.

[01:44:11] That we would sell more of these being super thin and a lot lighter. Because the original Folds were such... I remember I returned, I think it was the Fold 4 I had bought it. And it was like a gold brick in my pocket. It was just so heavy. But these new ones, you know, get around that. Thin and light is good if you can get battery life, right? Yeah. Yeah. And the Fold 8 has actually like really, really good battery life on it. But I'm curious if people are going to carry around the big Apple Pencil. Because don't forget the Fold had a stylus that would clip onto the back.

[01:44:40] You can carry it with you. This is going to be carrying a big pencil. Well, you're going to put that in your pocket. Put it behind your ear. Like an accountant. So, I'm glad it's there. I'm not trying to dismiss it. It's a great accomplishment. I'm just like, I don't know many people are actually going to use it. It's aspirational, I think, in so many cases. I remember when the first Samsung Note came out at CES. They hired a bunch of artists all over Vegas who had easels set up.

[01:45:09] And then they had their note on it. And they were drawing on their note to emphasize, I guess, that you could do that. And I bought every note that came out. I never used it. What was the one you got stuck? Yeah, that was, I think, the note. Oh, what did you put it in the wrong way? Put it in the wrong way. Yeah. Well, the sad thing is I was reading the story on Twit about how if you put it in the wrong way, it would break it. So, I said, you mean like this? And yeah, it broke it. Crazy.

[01:45:38] But the real point is I didn't care because I never used the damn thing. You were doing, was it JerryRigEverything before JerryRigEverything? I was, yeah. If I'd had a box cutter, I would have scored it. No, I think really, Jennifer, what I'm saying is I know it's a downgrade. I bet I'll be perfectly happy with the pictures I take with a duo, that it won't matter.

[01:46:06] Like, intellectually, I know it's a downgrade, but the pictures are going to look fine. I take sometimes, because I'm an old man, I take pictures with my iPad. What's this going to be like at concerts with everyone holding their duos up? Yeah, please, folks, do not bring your duo to the concert. To Leo's credit, I mean, computational photography is pretty good. It's everything these days.

[01:46:32] I mean, Google, the whole pixel line is built on this concept, right? But like, I have the Fold 8, and I think the cameras are fantastic. And you can argue like they're not even substantially different. They made the wide angles now 50 megapixels, so it has a different sensor in it. But the quality in it, like the ISP that Qualcomm did on there, along with the new algorithms, improved it in such a way that even I noticed it. But like, I don't really take photos even at nighttime and be like, oh, God, what have I done?

[01:47:01] It's not that dramatic. And I know Apple's always done well with photography. I think people will be totally fine with the cameras. I think there are other considerations in getting a duo, but I don't think the camera should really be one of them. I think most people will be pretty fine with it. I predict I'll be okay with it. And I think having their real estate will be nice. Yeah, I'm so excited about the real estate. They've done some really interesting things with the camera. You saw they have a little kid video that will play on one side when you're taking a picture of the kids. The kid looks at the video and smiles.

[01:47:30] They have, you know, with selfies, you can put yourself in the, or you could take a picture of somebody and have yourself in the. They're doing some really, I think they're doing some creative, interesting things. We'll see. Surprisingly, they didn't do the thing Samsung does. So Samsung allows you, when you open it up, you can switch to the cover display to be your viewfinder. And it disables the interior display. Then you can flip it around and hold it up and you can do a selfie. Don't they do that?

[01:47:57] Well, they do it, but they don't turn off the inner display for some reason. Oh, so it's still there. Yeah. Yeah. Like, I'm sure they can kind of fix that, you know, the, and the whole like cover display showing you Snoopy or whatever. Like, I know, I think Google has done that before the Razors have done that. So, but yeah, they, you know, I know a lot of people are giving that like saying, oh, Apple absolutely destroyed everybody. It's not so much that they had the last seven years to basically see how everyone has done foldables. It's very, and they just sat there and they took notes and they bought every single one. They broke it all down.

[01:48:27] They looked at the hinges and they just went to the drawing board and like, all right, we're going to amplify this. So yeah, their gen one looks fantastic. But I also would be disappointed if it was anything less than what they showed because, you know, it's like, well, you had seven years to watch everybody. Like, you know, to Samsung's credit, it's way riskier to be a gen one of a completely new form factor. And so like. Samsung has taken all the arrows in their back while Apple sat back and watched. And now.

[01:48:54] I really like the, I like the standby mode. Me too. I'm so excited about that because I use that all the time with the phone I have right now. What, so you use it like a clock or. Well, you can, yeah, you can put it. So book face down and yeah. Like tent mode. Tent mode. And the front will be the standby and they've adjusted the standby technology now. So it doesn't have to be plugged in because you used to, it has to be, used to have to be charging.

[01:49:19] And now you can just pop it on the side and it will go automatically into standby mode and be a great bedside clock. I'm constantly on the look for the like perfect. That's where Apple gets some credit here is that they've been a little bit more thoughtful about the different usage of positions and how to leverage the dual screen and foldable type concept. Right. Like for instance, they show the apps running side by side. This was taken directly from Microsoft, right?

[01:49:45] This is how the Surface Duo operated was this idea of like you run an app on each screen and when you dismiss one, it only dismisses that one, not the other. And so Samsung really hasn't taken advantage of this stuff. And of course, there's the fancy animation when you open and close it, which listen, I know it looks great, but. It's really cool. But as someone who just looks at technology as a tool. Yeah. I'm like, okay, but it doesn't actually do anything. Yeah. It made me roll my eyes a little too. Yeah, but I get it, right? This is what Apple does. That's Windows versus Apple right there. Right.

[01:50:15] Yeah. It's selling the experience and you're delighted when you use it. And I get it. I'm not trying to dismiss that. I'm just saying as myself, I look at things as tools. My first thought, Daniel, was how much battery is that using? Is that going to be a performance impact because it's doing that goofy little thing? Can I turn it off? Yeah. But it just goes to show you that like Samsung didn't do those kind of things and they had all these years to do it, right? So, you know, but this is the competition.

[01:50:42] Of course, Samsung's going to come out and, you know, someone's already created that effect on Android on the fold, but they haven't been able to build it into the launcher yet. Oh, I'm sure within an hour, somebody at the Samsung said, oh, let me see. Oh, yeah, we can do that. Oh, yeah, we can do that. Exactly. It was cool though. It caught my eye. Did we, Mike and I did the play-by-play for the event on Wednesday, which was unfortunately because of Apple's stinginess we had to do privately in the club.

[01:51:10] But Mike, I don't remember, did they say it does MagSafe charge? Yeah. They didn't say it during the event, but we found out afterward, yes, that it does. But the MagSafe looks a bit like, yeah, I mean, there are, yeah, it's a bit sort of bottom heavy. I don't know. I've seen a few creators already have shown some, like, because some of the accessories- Like it's hanging off the thing? Yeah, your wallets won't fit. You have to have a whole new line of accessories.

[01:51:40] It's a little bit lower than you might expect, but it'll still work. Yeah. So your accessories will all still work, but yeah, there is, and there are some new ones that have been developed, designed by accessory manufacturers that people- Well, there's an opportunity, right? Yeah. You could have it open and have it be a giant clock, and there's all sorts of stuff you could do. I am curious. So one criticism Samsung gets is that they didn't build in MagSafe into the phone itself, right? So it has the wireless charging, but not the magnet.

[01:52:09] But I always dismiss this because when I'm buying a $2,000 phone like this one, I do put on a case on it. So even if it's just a thin-borne, and it has the MagSafe built into it. But who's going to be buying $2,000 phones and not putting a case on it? I'm curious if you guys are- I already ordered my case. I've already ordered a leather Nomad case. Yeah. Exactly, right? So I always found this to be a weird complaint. It's like, because we're all going to put cases on our phones anyway. Same with the color, right?

[01:52:38] It's always fun with the colors. It's like- Black or white? Who's seeing it? No one. Yeah, who's going to see it? Doesn't even matter. Lisa, I said, what color do you want for your- Yeah, I like the orange. I do. The Aperol Spritz. But Lisa- It gets me very- I'm very popular with all my son's friends. They're like, you have the new iPhone. Now you're going to have to have Wild Cherry. Are you going to order the Burgundy one? I'm sticking with my- You're not even buying it. I'm a smart home person. I'm not a phone person.

[01:53:05] This was my first upgrade in like three years, I think, when I went to this. And that's how normal people operate. I got my whole family the whole- We all upgraded. Nice. So, and the kid- My son and my husband got the 17. So, and I was sad to see that we don't get the 18, the standard yet. Because they just laid AT&T fiber in my neighborhood. So I'm about to switch over. So jealous. I'm so excited. Is it symmetric or is it asymmetric?

[01:53:35] I don't know. I haven't got all the details. I did it while I was away and my husband was like, they kept knocking on the door and I was like, my wife has to do this. I can't do it. That's such a reverse of how it is in most households, right? That's great. I love that. Sign us up for a five gig plan. So I saw this, Micah, these beautiful Spigen cases that make it look like a MacBook Duo. I mentioned that Microsoft had done a Duo phone, but actually Apple, it's a trademark.

[01:54:03] They did the MacBook Duo years before the Duo phone. So I don't think they have any issue. I love these cases. This was a Spigen social post. Are they going to make them though? I'm not sure yet. They have not officially announced it, but the way that they're interacting with all of the influencers about it leads me to believe because they did the wonderful line of sort of old school Mac looks that I ended up buying every single one of those accessories.

[01:54:32] Yeah, you can get a kind of, I wonder if it ages from beige to whatever. Like yellower. Yellow. So far it hasn't, but they feel so good in the hand. I love it. Yeah. Yeah. Well, this is all going to be very interesting. We're in, we're in new phone retreat is always, uh, you know, it used to be until AI came along the single most interesting story of the year. Fortunately, thank God that AI has come along.

[01:55:00] Well, speaking of AI, have you, I mean, I'm sure you guys talked about this in the show. Speaking of AI, where are we, have we? But Siri AI gets unleashed on everyone here next week. Um, it'll be interesting. I mean, I have been very impressed when it can do things that I really wasn't expecting it to be able to do. Because our standards were so low. Yeah, I think that's true. I mean, it's not that impressive. Just the fact that it works is impressive, really.

[01:55:24] Well, like, I mean, I think I was at a hotel in Berlin and I said, um, I need to get to the conference hall and then the airport and my, um, tell me what time I need to leave. And that was the only context I gave it. And it got my flight. It got where I was. It knew which airport I was going to. It told me exactly what time I needed to leave my hotel to get to the conference center. Oh, that's what you want. Yeah. I mean, it was, and that would have taken me, I mean, maybe five minutes to Google it

[01:55:53] and figure it out. That's what it should do. But it just did it. And I was like, okay, this is, this is useful. This is the type of AI people want to use. It has caught me off guard with its capabilities. I, because I've talked before, it's was sort of a abusive relationship where in the past it had disappointed me so much that I just stopped using it, stopped trusting it. And so even with improvements that have happened over the years, I still sort of limited my interaction with it. And I thought, I'm going to open my heart to you again. I'm going to give you another chance.

[01:56:23] And so there'd be times where I'd be going, this is something I would normally just do myself, but I'm going to see, I'm going to see if you can do it. And I have been impressed with it, particularly in the car where before I just did not use it. I, one day I said, you know, I'd really like to get a slushie and I wonder if there are any places nearby that sort of specialize in that. And so I was able to ask it and then follow up and say, are there any close to where I am right now?

[01:56:52] And just those little follow-ups and the sort of details that require pulling in more information. And I, I've, I've been impressed. Well, there you go. There you have it. Our relationship is on your phone. And do you feel like Apple's being forthright when they say it's private and all of that stuff? We trust them more than we would, let's say, meta, right? I mean, I signed up for meta. So who do I trust? Well, I'm going to deny it. So I'm like, yeah, we, we, we, we want to test.

[01:57:21] I think given the, the bending over backward, we just talked about with how it's doing this Siri rewind stuff. Yeah. I do trust that the company is doing a better job than others, whether it is about using that as a means of selling more devices, or if it's an actual true core value of the company, I don't think that matters. I think that if you, if, if, if privacy is part of what the company uses to sell its stuff

[01:57:50] and it holds to that, then good. Yep. Let's pause for a moment and we will be back with our panel. It's nice to have some sensible people on the panel for a change. Let me tell you, uh, Jennifer, the comics always do that audience last night. They were terrible. I'm so glad to be in Philadelphia today. Jennifer, that is in too. I actually always love having you on. She is so great.

[01:58:18] Uh, we'll talk about her travails. I'll explain to the dog with the weed whacker on its back. I should probably explain that since I started the show with that in just a little bit. Uh, you'll find her at the verge.com. Daniel Rubino, senior. No, I'm sorry. Editor in chief of, uh, Windows central. Great to have you as always. We'll talk a little bit about patch Tuesday. It was a record breaker, uh, in just a moment. And Micah Sargent host of, well, many great shows iOS today. I'm sure you'll be covered.

[01:58:48] So let's see. iOS 27 comes out tomorrow. Is that right? When is it coming up? God. I know I have it already cause I'm in the beta, but I, what is it this week? Pretty soon. I don't know that they've officially, have they given a date, but I think it's Monday. Yeah. Monday, September 14th is the rumor. I think it's tomorrow. Okay. Oh yeah. There it is. Apple's updated its website now available starting nine 14. So this week is all going to be about coverage of Apple. 27 is here. 27 releases.

[01:59:17] So watch, uh, iOS today for that with Rosemary Orchard and Micah, and of course, tech news weekly on Thursday and coming October 1st, hands on a eye. Our show. Let's talk about privacy. Our show today brought to you by delete me. Holy cow. There is something we have here in the United States. We're so proud of. We're so proud of the data broker industry. They make so much money. There's such a success story.

[01:59:44] There are more than 500 of these companies with more cropping up every day. Cause it's so profitable. What they do is collect information about you. They collect it from all sorts of sources, your internet service provider, your mobile carrier apps on your phone. You'd be amazed your TV set. They collate all that information to build a dossier about you and your family and your company, everything there is to know, and then sell it on to anybody who's got a buck 50.

[02:00:14] And if you're a business owner, this should, I mean, everybody should be concerned about it, but business owners in particular, you're putting yourself out there. It's kind of part of the job, right? It's part of marketing, but here's the uncomfortable truth. Promoting your business leaves you and your team exposed. 90% of business owners, business owners have their home address easily discoverable online. 90%. And the average owner has more than 600 pieces of personal information just sitting on the open web.

[02:00:44] We're talking your personal email, phone number, home address, details about your home. Now, let me tell you why that matters. Yeah, it's bad enough for we individuals, but for business owners, it's a real risk because who's buying that information? Well, yeah, marketers, intelligence agencies, foreign countries, foreign governments. But the ones you should worry about is hackers because they use the data about you to make

[02:01:12] their phishing attacks more effective, more targeted. They have your real details, so they don't sound like strangers. And I got to tell you, it's worked. It's worked on us. We got an email from a client, somebody we deal with regularly with what they call an RFP, a request for proposal. It went into our, I think our continuity department. Actually, first it went to Lisa.

[02:01:37] And because I have some software on here protecting her, she couldn't open the request for proposal. So she sent it on to the continuity, her staff, and they opened it. And it looked completely reasonable. We have to respond to those because that's a client, right? I mean, somebody's asking to buy ads, so we respond to it, except that it wasn't a client. It was somebody posing a client because they knew who our clients were. They knew who our CEO was. They knew who our continuity department was.

[02:02:06] They knew our entire org structure. So they sent a crafted phishing attack that was designed to be something we couldn't not click. When we clicked it, it popped up what looked like a login to our Google Workspace account. That happens all the time, right? So we just typed it in. I'm not going to say who, but just typed it in. Turns out it wasn't. It was a man in the middle that looked exactly like the Google account. This happened back in the spring.

[02:02:32] And they had given the bad guy everything, including the two-factor authentication, because they were sitting there waiting for it, and they immediately got into our Google Workspace. Now, the good news is we caught it. We got rid of it before any damage was done. But that was really scary. That's as scary as the other time when our staff got targeted text messages from the boss, except it wasn't from the boss, saying, buy us some gift cards and say, I'm in a meeting. I can't do it right now.

[02:03:01] Look, because they have your real detail, they don't sound like strangers. These bad guys sound like clients, partners, your boss. That's why attacks using verified personal information are five times more likely to succeed. And the average incident costs a small business more than $120,000. And one in four businesses will be impacted this year alone. It's happened to us, to my knowledge, twice this year alone. That's why we signed up for Delete Me immediately.

[02:03:28] Delete Me reduces exposure by up to 95%. How do they do it? They do it by removing you, your management, your employees' personal information from hundreds of data broker websites. And that starves hackers of the fuel they're using to build these highly targeted attacks. Very important. It's not a one-time thing. It's a subscription. And you need that because, yeah, you could pull down the information today, but these data brokers, they're cockroaches.

[02:03:57] You can't kill them. They'll start rebuilding the profile or they go bankrupt and change their name and just start over again. So Delete Me knows about this. They know where to go to remove your data and do it again and again. Plus, you'll get regular privacy reports so you always know what's been happening, what they found, what they removed. Fortune 500 companies, government agencies have been using Delete Me for over 15 years for exactly this reason.

[02:04:26] And now that same enterprise level protection is available for you and your small business. It's why we use it. It's why you should use it. Protect your business. Protect your peace of mind. Go to joindeleteme.com slash twit-biz to start protecting your business with Delete Me today. And if you use that link, by the way, you'll also get a free year of social media protection for every seat you purchase. It's joindeleteme.com slash twit-biz.

[02:04:54] Joindeleteme.com slash twit-biz. And we thank them so much for protecting us. Thank you, Delete Me. Great, great partner. Great sponsor. And we're very happy to have their service. Patch Tuesday, I was doing security now with Steve Gibson and the information hadn't come out yet. And he said, we're going to watch this with interest because either it's going to go up or it's going to go down. The number of fixes.

[02:05:22] His theory, and I think Steve's smart, is that as more and more of these patches get fixed, eventually there'll be a point at which the number of patches, they'll go up, up, up, up, up, and then they'll start going down. And eventually be zero because everything will be fixed. Not this week. This week, Microsoft Patch Tuesday released more fixes than double the number of fixes they released all last year.

[02:05:54] 972 vulnerabilities patched, 112 of them high severity. Were you surprised, Daniel? No and yes. I mean, obviously the number is high. And it was a big increase. We should say that's good. It's good. And I'm not surprised because of, obviously, the tools that they're using, right? So it's not like that Microsoft is finally taking security seriously.

[02:06:20] It's just that now there's AI tools that help them go through and find these exploits ahead of time before they get out there. So this is where, you know, AI has a more interesting role in terms of software development. It can be both used as a sword so you can probe systems and find attacks and people can use that to exploit. Or a company like Microsoft can do that to themselves and try to find the exploits before they are public and patch them.

[02:06:46] And, you know, it's doing the job of more and more humans, which I think is actually a really good thing. So, you know, this is a good thing. It's making the operating system more secure. It's doing it before these exploits were made public in the first place. And it goes to show you how this technology does improve. And to your point, yeah, you know, in theory this stuff could drop down. Of course, they may release Windows 12 a couple of years from now, in which case it'll probably go back up again.

[02:07:13] But they do have a lot of old legacy code and it makes, you know, trying to patch this stuff very challenging. So it's nice to see the technology kind of working in the favor of them. But yeah, it'll be interesting to see these numbers go up or at what point they start going back down again. Are they using, do you know, are they using AI? I mean, they must be. Yes. Yeah. Yeah. So this is really a success story in some ways for things. I believe so. Mythos. Yeah.

[02:07:39] Are they using their own models, Microsoft models, or are they using Mythos? What are they, what are they using? That I don't know the specifics of, but I wouldn't be surprised if it's. No one does there except the people inside Microsoft. Right, right. So, but I do believe that, yeah, they're using these tools and I think it's helping out a lot. That's why they're discovering so much because they know if they don't do this, companies are going to be using AI themselves to probe and find the exploits in Windows. Bad guys. In bad guys, yeah.

[02:08:08] So, you know, they're trying to get ahead of the situation here while they have these frontier models at their disposal. Right, right. So, and the other thing I would say is I said sometimes people say, oh, you see how buggy Windows is. But it's a big code base and it's always good that they find these and patch them. That's just a good thing. And buggy, and I want to even, you know, buggy implies bugs, right? Like errors and, you know, things crashing.

[02:08:37] And that's not what this is. A security exploit isn't a bug in that traditional sense. It's a bug in the more general sense, right? But these aren't, they're not patching things that people have noticed. You would not encounter these. Right, right. Unless you're back. But they are also going to be using, yeah, they're also going to be using AI for, you know, quality of life things, right? We even see this in video games, right?

[02:09:00] They're using AI and video game development as well, not to, some companies are trying to steer away from creating the media assets for games. But instead they're using it for validation tests and making sure that the game performs in certain ways. And I think this is a strong case where AI actually does play a super important role that shouldn't be really controversial because it can do the work of many humans on a different scale.

[02:09:27] And whether you're an indie developer making a video game or you're a giant corporation doing one, you know, using AI to stress test and look for these exploits, I think could be very valuable. Yeah. Let's talk TVs. Gamers Nexus, a YouTube channel, did a very interesting thing.

[02:09:53] I guess we've always thought, except for a long time, that smart TVs were spying on us. Yeah. I don't think any of us thought they were doing it at this level. Not to mention a company like LG, right? That's the biggest problem. Yeah, like Vizio, it's kind of like Roku, you know. Come on. Yeah. Yeah, Roku, you know, it's like, but LG is like... Or Tele, the one that actually does spy on you and tells you. Oh, that's right. You could buy a free TV.

[02:10:22] It was a free TV. Buy it. You don't buy it. It's a free TV. Yeah. And it wasn't so much spying on you as putting an ad at the bottom, right? Like it was a scroller. Both. It had a camera that watched you to make sure you were watching the TV. Are you watching the ads? How do you feel about them? So what is the LG doing? Gobbling. It's a little mixed, I guess. Yeah. I mean, if you go by what LG says, they're not doing anything wrong. But if you go by all the other independent analysis, there seems to be a lot of recording

[02:10:52] of the usage of microphones and, you know, data being sent constantly. The thing is, to their credit, you know, a lot of the data is encrypted. So it's not like you can necessarily see what's being sent. You could just see where it's going to and the frequency at which it's happening. I'm looking right now at an LG OLED screen. It was the best TV ever. I mean, it's really good quality. Is it recording what I'm saying right now? It's off.

[02:11:21] The theory, well, the argument that they made or the point that they were saying, the LG was saying that it's similar to the voice assistants. So the Alexas and the Google Homes, you know, it's listening for the wake word because you can control it with your voice. Right, which I don't use. And you can either use the word, you can say it out loud ambiently, or you can press a button on the remote to trigger it. So you can turn. So it's listening in the same way at Amazon Echo.

[02:11:51] At Amazon Echo. Well, yes, that's how they characterize it. That's how it keeps that. But the discovery was that it's constantly sending. So they're saying the discovery is that actually it's constantly listening. Because the Amazon Echo doesn't send anything to Amazon until it hears the wake word. It's listening locally. Or it thinks it hears the wake word. Right. Yes. And then it sends whatever it follows to the servers and processes that.

[02:12:18] And then, but it's not sending anything. And if it hears the wake word, it mishears the wake word, then it in theory immediately deletes anything that hits sent because it says, oh, this was misheard. This wasn't the wake word. But yes, that's the way LG's TVs are supposedly working is just by the same way that your smart speakers are, because it has a built-in voice assistant like a smart speaker does.

[02:12:43] But the data implies that it is sending recordings constantly. And LG is disputing this. It's also saying that it was sending content, like what content you were watching using. Yeah, it's like a Shazam feature. It tries to figure out what you're watching. What you're actually watching. So they use, so Gamers Nexus used Wireshark to watch the network traffic. Yeah. They also looked at the TV firmware.

[02:13:13] They also looked at the memory in the TV. They said that LG TVs are, quote, capable of identifying your IP address, geographic location, the name of every Wi-Fi network in the vicinity. But everything with Wi-Fi can do that. And signal strength, channel numbers of neighboring Wi-Fi networks in range, and other devices on the network, even if not connected to the TV in any way. And the internal IP addresses of other devices on the network alongside others. I got to tell you, your laptop can do the same thing.

[02:13:43] Anything with a Wi-Fi radio could do that. And I think the concern people have here is that they're not expecting their TV to be able to do this. Like, you're right. Lots of devices can do this. But you know that. And you're aware of it. You're used to seeing a prompt that says, can I have access to your local network? Can I do this? And this is a little bit more buried.

[02:14:05] And in particular, I think the issue here is that why is it that even when it's offline or on standby mode that it's doing this? We were able to capture, Gamers Nexus said, microphone audio while the TV was unplugged from the network. So it's not even on the network. The audio was reportedly stored locally in plain text. So the theory being, as soon as it connected to the network, it would just get sent right up.

[02:14:32] But again, because it was encrypted, the Wireshark couldn't see what was the content of the packets being sent. LG's response is... I think you see here. LG smart TVs do not, and I want to underscore this word though, they did not continuously record or transmit users' conversations.

[02:14:56] Speech-to-text processing begins only if a user activates a voice interaction through a supported wake word feature by pressing the voice button on the remote control. Audio is processed locally on the TV for the wake word. And if no wake word is detected, it is not converted to text, stored, or transmitted. That's how Alexa works. We know that, right? Yeah, it's the same kind of mechanism. Voice recognition results and related technical logs may be generated as part of processing a voice command.

[02:15:23] These records are associated with specific voice interactions, do not indicate continuous recording. So, I mean, they're saying we don't do it, but they're not saying, well, they're saying we do it, but only in certain circumstances. It also uses something called automatic content recognition, voice recognition, and interest-based advertising. Those are optional. You can turn those off. And LG says they're not even on by default.

[02:15:53] Users can choose to enable those features and can manage or withdraw consent through the TV settings. Good luck, though, looking through the TV settings. So, to ratings credit, the site ratings, they have a video out right now, and they go through and they tested a bunch of TVs and boxes as well. Yeah, they all basically do it. And the trouble is they all call it by a different name. So, you have to kind of figure out what each one is called.

[02:16:21] Some have different – some, yeah, you can just disable, no problem. Others make you, like, reset the TV and go through the process again, and then you have to tell it no on all the various things. You know, so there's going to be different levels here. And there's, yeah, questions about what TVs – some send more, some send less. You know, it seems to be that if you want the safest approach, you use – get a smart TV. Don't let it connect to the internet.

[02:16:46] But then you use some sort of box to connect to, you know, Apple TV or something and use that as the interface. Richard Campbell says that if you put a pie hole in, you can control what data is sent. You can still get your TV updated, but you can say, don't send anything back to the home office. Yeah. And so, it is possible to connect and then filter. Yeah, because – A pie hole is – I know, it's beyond most people. I know.

[02:17:11] What they're saying here is, yeah, we don't do – we're not sending this – we're not recording everything you say and we're not sending it. But they are recording some things and they are sending some things. And what are they then doing with those recordings? And, you know, it's all feeding the advertising. And that's obviously where – and this is – LG is not alone. Many TVs do this and we've known this for years. But LG definitely got – you know, what they – the concern here is that they're doing way more than they say they're doing.

[02:17:39] I mean, they are recording way more than they say they are recording based on the evidence that was uncovered versus what they're telling us. And you can turn these things off. You can disable them in settings. But are they really disabled? Exactly. I mean, how do you actually know? We can't trust that now. And this is where – I mean, to Dan's point, and this is what we've recommended, is, you know, you buy your TV. And there is actually one – I think it's Panasonic. We wrote an article about this recently that we thought they accidentally launched a TV that you don't have to connect to the internet.

[02:18:09] But it seems like it was actually on purpose. So you can now buy a TV that is dumb. I want a dumb TV. It's a good TV that has a good screen, you know, and not – and it has no Wi-Fi requirements. Because you can connect to the internet, but you do not have to. Because this is the problem. A lot of these smart TVs, they just will not work at all. Even if you plug a box in, you still can't get them to work without connecting them to the internet.

[02:18:33] So, yeah, that's the best solution if you're concerned about this is stick an Apple TV box on one of these things. Even just get – some people suggest getting a projector because that doesn't have any of the hardware. Well, it depends what kind of projector you get. And also you spend an awful lot of money on a projector. But, yes, there are solutions around. And I think we might start to see more manufacturers come out with TVs that aren't smart because there are people that really do want them.

[02:19:01] They don't want to have their TVs connected to the internet. And two points to this too. When the first 4K TV was showed off at CES, its retail price was $20,000. That was the very first 4K TV. Now you can get them for $200. One of the reasons why TVs are so affordable is because of this model. It's the same thing with PCs and laptops, right? Where they'll get subsidies through pre-installed software, which angers everybody.

[02:19:31] But you can literally just right-click and uninstall it. But it's one of the reasons why the laptop prices can be competitive. And that's what you're seeing here. It's like I was just – I'm recently looking for a new TV. And I'm shocked at how low some of the prices are for some of these devices. So that's the tradeoff that you're getting for a lot of this. The other one is even though if you can turn all this stuff off and out down, all that, that's good. But there is the – this is a vector for attack.

[02:19:57] John Kiriakou, who worked for the CIA, famously went to prison for talking about the torture program, has talked about this. The CIA can easily hack someone's TV, turn on those microphones, and record everything, even if the TV is powered off and all this kind of stuff. Oh, that's a good point. Yeah. So like these are just different attack factors. So that's the CIA. And I get a lot of people are like, well, I'm not doing anything.

[02:20:22] But the point is that there's no – especially with AI, that other nefarious actors couldn't figure out how to do this. And they probably already are, right? So this is just something for people to think about. How much of this, Daniel, is related to – I suspect that Gamers Nexus did these tests because last month LG Monitors started downloading McAfee ads on Windows PCs. Is that right? Did I get that right?

[02:20:50] Even Microsoft stepped in and said, whoa, all right, guys, you got to slow it down. Because they were using Microsoft's Windows Update feature to install this. Yep. It was sort of like a help around. Microsoft made them stop. Mm-hmm. Yeah. Microsoft made them stop. But the fact that they did it in the first place kind of opened the door to what else might they be doing? It's like every company in the world, right? They're going to push to see how far they can get away with something. And then like, all right, you got us. All right, we won't do that. It's a bridge too far. And no one said anything.

[02:21:20] Not only would they have done it, other companies would have started doing it too. I have to correct myself. It wasn't Panasonic. It was Vizio. Vizio. Vizio. Vizio made a great dumb TV by accident. By accident. It might have been by accident. Yes. So there are options out there now, which there weren't for a long time. I've seen people come up with a solution of just like buying like panels. You know how like companies buy like panel, just a panel to run ads or something.

[02:21:46] And just plug an Apple TV box into that because it has gotten so frustrating for people that don't want to connect their TV to the internet. And there aren't many dumb TVs out there anymore, sadly. A tip for that though. Another way to get it. I got good news and bad news. The bad news is that firmware we put in the TV, it doesn't work. The good news is they're selling like hotcakes. Yeah. Everybody seems to want it.

[02:22:13] Another place to get a dumb TV is store signage. You know those signage things on? Yeah, that's what. Right. That's what I was talking about. That's what you're saying. Get like the menu board from your local sandwich shop. Yeah. It'll have bogeys burnt into it, but no spyware. There's a lot to be said for, you know, it's probably a niche market at this point, but this push for dumb technology again, right?

[02:22:38] You're seeing it, especially in the auto space, a lot of people are really upset with the way cars are going. You know, what I call the Tesla-fication of cars. There's, even though it's like, in my opinion, Tesla gets a lot of the software stuff done right, but the other auto manufacturers are having a lot of issues, I would say, trying to get it stable. A lot of people are just getting it, like they just want to pick up. They don't need all this giant infotainment system connecting to the internet. Five apps got to do this.

[02:23:04] And so there is this kind of market, I think, for getting into dumber technology where things are just not connected. How far that goes, I don't know, you know, because it is convenient and ultimately that's what people kind of want. Well, and you get the fast TV channels now, like a lot of these. The free TV. The free, yeah. The free ad-supported TV services, which have gotten surprisingly good.

[02:23:31] I have to say, though, they must have worked hard to get that acronym. Oh, these things are fast. Yeah, exactly. Free ad-supported TV. It's like, I feel like we had this at one point. How to get a TV? Oh, yeah. It's what we called network television back in the day. I pretty much watch regularly the All Cheers channel. It's All Cheers all the time? Yeah. I love those channels. They're the best. They should have called it Hello Norm. Yeah. That's funny.

[02:23:59] So you're saying they're pretty good, Jennifer? Like what? Like All Cheers all the time? So the fast TV channels, there's all sorts on there. I mean, you can get like the NBC News channel on there. Okay. And like not your regular over the air. So for us older people that grew up with ad-supported television, it just feels like home. Oh, yeah. Because I dream of TV has those pauses every five or six minutes for an ad. You might as well stick an ad in there. You guys don't use Pluto TV?

[02:24:30] And you can have like All Cheers or All Baywatch or All Friends. When Amazon started showing ads and said, but for $4.99, you could turn it off. I immediately clicked the button. I just... So I would actually go try Pluto because like they have channels dedicated for every sitcom from the 70s and 80s and 90s. That's an interesting idea. They also have like thousands of movie channels. They'll have like drama. They'll have horror. They'll have action. How often do the ads come up? Are they a lot? It's like... No, no. It's...

[02:24:59] Knowing that sometimes the ads, they don't have an ad. So it'll just play some like nice music. And it always has a counter in the corner. It tells you like 30 seconds left or something like that. And all these shows... My pet peeve is... Bad breaks. All the shows... In the early days of internet streaming TV, there weren't enough advertisers. So you'd seen the same ad eight times. It still happens. I hate that. Yeah. I know. I don't want to see that ad again. Yeah. But like the commercial breaks aren't... It's not like every two minutes.

[02:25:29] It's not as bad as it used to be. No, no. It's... I find it acceptable. Plus it gives you that chance to go to the kitchen and get a sandwich or whatever. I was going to say, if it's in the background too, if it's just background noise. Yeah. It feels very... There is still something to be said about browsing and like, oh, look, this movie's on right now versus... Versus... Yeah. It's like the old days. Picking a movie. Yeah. Oh, boy. You might have just convinced me to check out Pluto. I'm going to check it out now. Is 2B also that or no? That's something else. 2B. Yeah. There's a few of them.

[02:26:00] Samsung has it. Flum and Binggong. I feel like 2B is kind of cringe though. I don't know. If I told someone I was watching... 2B? 2B. I'd be going... 2B. It's got a... It's a cool... That's a cool chime. You must be a green bubble. You know what I watch like every single night? It's because there's a dedicated Mystery Science Theater 3000 channel and a Riff Track channel. I watch... All I watch is Mystery Science Theater. And so it's like... It's just on 24-7. See, I like that kind of a thing. Oh.

[02:26:27] Well, when you're in the mood, you know, you can always go where everybody knows your name. Or you can always... You know, you just know that there is... You know, it used to be we would say there's an I Love Lucy episode playing at any given time somewhere on the planet Earth. It's like that. You can always watch, you know... I just like... It's just background. And so you just want something on the background, you know? Yeah. Hello, Norm. My problem is it keeps popping up on the background on my fridge. Sorry. On your fridge.

[02:26:57] Do you have one of those... My washing machine always playing at a Lucy. And it has... So Samsung has its own Samsung TV Plus, I think it's called. What, do you stand in front of the fridge watching TV? I don't. But occasionally it just starts playing TV shows on my fridge randomly. Is that what they think people are going to do? Like they're just going to stand there with their mouth open in front of the fridge? Like Wally. Well, it's... You know, you have a TV in your kitchen often when you're cooking. If you're cooking. Yeah. And just have something on in the background. Oh, okay.

[02:27:27] Oh, that makes sense. All right. But it's just, I didn't ask it to. It just pops up. That's what's concerning. That's the way it's been. I don't know. Is it scary? Like, what do you think is making that happen there? I know. I don't know. It does have a mind of its own. And it's weird when it just shows a really odd TV show that I wasn't intending to watch. I don't like that. I don't like that. No. I really don't like that. Yeah. I thought I would like watching TV on my fridge, but I don't. Only you would say that, smart home mama.

[02:27:59] I mean, people don't remember, but in the days of network television, a half hour TV show was 22 minutes long, so they could put in, what is it? 18... No, eight. Anyway, a lot of commercials. I'm so sorry. Could you be quiet? I'm watching the X-Files on Pluto right now. Thank you. Yeah. Yeah. That's an X-Files channel. Yeah. It's good stuff. I'm telling you. I don't. Search for Spock? I can't wait till I'm in the old folks' home, and I could just leave that on all the time.

[02:28:28] Oh, it's Mulder and Scully again. There's a Star Trek channel, for God's sake. I know the truth is out there somewhere. You're watching This Week in Tech. Our show today brought to you by Scribe. Scribe! Scribe! Think about how much your team has figured out on their own shortcuts, fixes, better ways of doing things that have never actually been written down anywhere. That's the gap Scribe was built to close.

[02:28:58] And that's what today's sponsor, Scribe, does best. Scribe is a specialized intelligence platform trusted by 94% of the Fortune 500. You just do the process as you normally would, and Scribe captures it in real time, turns it into documentation automatically. There's no manual writing, no manual screenshots, no starting from scratch every time someone new joins the team.

[02:29:24] Scribe automatically redacts sensitive information, names and account numbers and emails. As an admin, you can enforce this across your entire team, and nothing is published without human review. Feel confident nothing slips through the cracks. Anyone following this process can launch real-time on-screen guidance, showing them exactly where to click inside the actual tool so critical processes get done right every time. And that context can power your AI agents too.

[02:29:52] Most AI was trained on the internet, not your company. So Scribe helps agents operate on how work actually happens, not guesses. To see what Scribe could look like for your org, head to scribe.how.twit. And mention Twit for your first month of Scribe capture free on select plans. That's S-C-R-I-B-E dot how.

[02:30:18] Thank you so much for supporting this week in tech. It's really interesting how many of our advertisers these days are AI companies. I think 100% or close to it. It's a really interesting time. As long as we're talking privacy, LG TVs, the flock backlash is everywhere. City governments are being told, take down those cameras.

[02:30:46] Although apparently in many cases, they're replacing the cameras with cameras from Axon. So it's not flock anymore, but it's still watching you. And now retailers are putting it in. Ulta Beauty, which is a beauty store. We have one in Petaluma. Turns out, according to Rachel LaPointe, who is a Substack newsletter writer, turns out to be flock's biggest commercial customer.

[02:31:14] I mean, there's no license plates inside an Ulta, I don't think. There's a lot of shoplifting inside an Ulta there. I guess so. Yeah. I mean, didn't those flock cameras start like in Home Depot parking lots? I believe, I don't know. I believe it if you say so. I used to see them on there all the time because people would, because of all the stealing of, you know, high priced tools. But they're also in places that they should, feels like they shouldn't be like playgrounds. Yeah, that's a big thing.

[02:31:41] I've even seen some people have turned down buying a property because there's now maps that you can go check to see where the flock cameras are. Every time you pull out of your driveway, you're on camera. And so sometimes people will be on the camera and they're like, oh, I'm not going to buy this house because it's part of the angles on a flock camera. Maybe you should get this car, which is painted to, to make flock cameras lose their mind.

[02:32:07] And Bill Swearengin, he's the founder of Six Cyber. He spent the past year running 31 million tests, developing something he calls no recognition. It's a reinforcement learning model that generates patterns capable of defeating the algorithms inside flock cameras and other devices. He actually has a Kickstarter campaign going,

[02:32:35] no recognition that you could buy adversarial clothing. So here's a, I don't know, would you wear this? It's actually pretty striking. How do you describe your fashion sense? Yeah. Adversarial. So it's a, so I guess if you're on, you know, the, the, this will make the camera crazy. You know, I should just get one of these. You should. See what happens. Yeah. But it's just one of those things where it kind of kicks the can down the road. It's like, if your car looks like that, it's like, all right, we just now need to look for this really crazy. Look for the car with the,

[02:33:05] wasn't it like in a naked gun or something like that, where someone, uh, they, they, they cut and copied the letters and wrote a letter. So they couldn't trace it. And they're like, all we gotta do is find the crazy typewriter that did this. Like I, my, uh, my ex-wife never liked coming by the studio because there were cameras everywhere and she didn't want to be on camera. And, uh, Colleen Kelly, our, uh, chief engineer at the time wanted to build her a, a, a ultraviolet light ring that she could wear around her neck,

[02:33:33] like a necklace that would beam invisible light to you, but light that would blind the cameras so that when she walked into the studio, she was just like a glowing ball of light and be invisible. I don't think Colleen ever built it, but she, she worked at meta for a long time. Maybe, maybe she did. The no recognition Kickstarter also offers this, an elastic tube gator. That you pull on, uh, so that you aren't susceptible to, uh,

[02:34:03] face recognition. You may mock it, but notice it's sold out. He sold out of them. You know what else have as your face? A regular bandana. I mean, yeah. Well, that's true. Except that somebody might mistake you for an ice, uh, officer. Yeah. No recognition hoodie, 75 bucks. The flagship. I don't get it. You still see her face. I need some like hardcore proof. I need to buy one of these and go stand in front of my cameras.

[02:34:33] Just go. Hi. How do you know if it works? Well, I have. So, like I said earlier, my cameras, when I walk up in front of them, recognize me and they'll say, you know, just the pattern will make it go. So it makes the camera go crazy. Right. Like, so it's not that it's not, it's not that it's not seeing your face. It just turn, makes the camera. Watch. So you can see it's detecting a person. And as soon as the camera, for some reason,

[02:35:03] it just may, it's, it's like hacking the camera. That's of course, this becomes a, just having the pattern somewhere on you means. See you. Yeah. Well, that's what he says. I guess it's a tit for tat thing, right? At some point, I would imagine the engineers be like, okay, we'll try to quote around that. Right. Exactly. If you detect these kinds of patterns. Yeah. Like it's, I don't know. I love it that he's using AI to generate these. Like, what can't you recognize?

[02:35:31] And knowing the country we live in now, wearing that will probably get you stopped. Right. Like, well, what are you doing? You wouldn't be wearing that unless you're doing something wrong. So what's, what are you up to? But you know what? There is good reason people are nervous. 404 media has discovered the, the DHS department of Homeland security is doing pro predictive policing by analyzing Americans financial data. Of course. They,

[02:36:01] they analyze the data to find people to pull over and investigate. By the way, I, I don't know if that's legal. That doesn't sound legal. Like it violates the fifth amendment or something. DHS and other, uh, LEAs law enforcement agencies need to have a reason to pull you over. Right. Probable cause road behavior. But how do they get,

[02:36:30] maybe they can follow some long enough. You could find probable causes. You accidentally go over the line. That's a moving violation. Yeah. Yeah. Hmm. How are they using your financial data? What they think, Oh, they're so poor. They're likely to go. Yeah. Or something like, Oh, you're getting paid in cash or I don't know. I don't know. That's what I don't, we don't know. The predictive policing thing is scary. That's the minority report. I was going to say like minority report was an instruction book for law enforcement,

[02:37:00] which is like, all right, we just got to wait for technology to catch up. And of course that was going to happen. Which does a face recognition. They get a lot of trouble because their face recognition was a little racist. Shall we say? Now they're testing an AI tool that would let cops unearth your life online. This is from wired inquiry IQ, a previously unreported prototype tested a model from XAI to surface associates, social accounts,

[02:37:28] and other information about people identified through clear view. So I guess it's the next step after face recognition is, well, let's go analyze their social media and find out. And I bet it understands sarcasm. I bet it. Oh, it's so good at understanding the sarcasm that I'm doing right now. No, Which means you're in the database, buddy. I'm there now. I was already there.

[02:37:58] So it's the unreleased tool is designed to take details and investigator unearthed from a clear view search, a search of face recognition database, which we already know sucks. And then automatically fan out across the web, opening web pages, analyzing images and assembling what it finds into a profile containing possible employers, aliases, associates, and physical characteristics of a person. Police are investigating. Don't they already do this? I didn't like, I mean,

[02:38:26] I feel like it's just an extension of the Homeland Security, September 11th fallout, right? This is what they were already doing to try to connect dots. It's, you know, this is what happens when you, you know, prioritize fear in your society. Right. And, you know, not to, not that the Democrats don't do this, but Republicans are very good at fear. Everyone's trying to kill you. These people want you dead. These people hate you. So in order to stay safe, I'm going to allow you to take all the cameras around my house and use those in police investigations.

[02:38:55] Then everybody acts surprised. Like, what do you mean they're spying on me through my TV? Right. Well, but you're safe. It's going to happen. And you know what? It's true. It works. you know, our locality, our town, our little town, which has, I think a half dozen flock cameras published a report on how many missing senior citizens were found because of the flock cameras. Yeah. Well, that's great. I mean, literally the problem with all of this. Yeah. I mean, I grew up in London. We have cameras on every street corner and it's,

[02:39:25] yes. For that. It's yeah. And I grew up in an era when we had people putting bombs in trash cans and blowing up train stations. And it definitely helped stop things like that. But the problems you have are based on the government that you are under. Like, do you trust the government to use this technology responsibly? Yeah. When you combine it with deep fakes and the ability to like, say the government goes, we got you doing all this on video and they show video. It's just like,

[02:39:55] yeah, that's not me. And it's just like, you know, and we've seen the plenty of arrests by flock cameras where the police go to like a woman's house and they're like, we got you in front of someone's house, stealing this package. And, you know, it turns out, no, it was wrong, you know? So it's just like, it's a strange new world. Well, and then there was the Nancy Guthrie case where they had all the footage from the doorbell camera and then it did no good whatsoever. So, I mean, it's a double-edged sword.

[02:40:23] There are definitely arguments for how beneficial cameras can be in solving some crimes. Not a lot of evidence, but some evidence. And not, I think the biggest thing is that it's not really a deterrent. I think they've proven like the fact that there are cameras doesn't really stop people from committing crimes, but it does potentially help you find the person, but not always as has been proven.

[02:40:51] But having some information is obviously better than none. It helps police, but when it's abused, if the police abuse it, or if your government abuses it, then this is... There's all those cases, right? It's like the police, it's like, oh, they've accessed, they're tracking this person 45 times today. And it's like, oh, who is that person? Oh, it's an ex-lover. It's like, come on, man. Yeah. So that was pretty bad. Yeah. I mean, I feel like the police have like done the most damage to flock. It's just like, you know,

[02:41:20] where everybody here is like, there's the potential for governmental abuse. And here's the government out there, like literally abusing it every single day. And it's all just come out. And it's just like, it's like, okay, that's why people are reacting to this stuff. It's like, there's not the potential. They already are using it for like the dumbest things. It's just like, you know. let the terrorists win, but that's the point of terrorism is to terrify you. And the, uh,

[02:41:49] what do people do when they're terrified? They give up their civil rights. Yes. They, uh, they, uh, they lose their freedom. Uh, it sounds like the terrorists won. Sad to say. I think that is a lesson from nine 11. Literally everything did change from that time. And we live in fear now. Yeah. Kids don't go outside anymore. Yeah. And, and that's the thing. Kids don't go outside anymore.

[02:42:15] And yet the crime rates between now and then have gone down. It's safer in New York city than ever before in history. Yeah. And here in war torn Portland, it's very safe. Well, I may not like surveillance, but I am putting a camera in my toothbrush. We'll talk about that. Wow. I could have known you were going to buy that. The second they thought they launched it. I've got a flock camera in my toothbrush.

[02:42:45] We'll talk about that. When we continue our last segment, your cavities are illegal. You've been very patient. Jennifer Patterson, Tui, Daniel Rubino, and Micah Sargent. Wonderful to have you on the show. I really appreciate it. Our show today brought to you by Bitwarden. I do. I have to say when it comes to trust, there's no better way to get my trust than to be open source. And if I'm putting my passwords and all my secrets in there, I want it to be open source. That's why I trust Bitwarden, the trusted leader in passkey, password,

[02:43:14] and secrets management. More than 15 million users now. Congratulations, Bitwarden. Over 180 countries, 80,000 businesses. Bitwarden is dedicated to helping individuals and organizations protect their digital lives with trusted open source security. And I have to say, I interviewed the founder of Bitwarden, Kyle Sheeran. I was so impressed by the guy. A couple of months ago.

[02:43:40] And he reassured me that Bitwarden's absolutely dedicated to the idea that everyone should have access to strong security. Fully committed to forever offering a basic, and let me underscore this, free password manager for individuals. Now, they support themselves with paid plans for families, which I pay for, for teams, for businesses. But they believe it's so important that individuals have this protection

[02:44:09] that they will always offer that free version. And it is a full featured free version. Unlimited passwords, unlimited devices, passkey support the works. And getting started with Bitwarden is so easy, easier than ever. They have a new direct import option, which means you can move your existing passwords into Bitwarden with just a few clicks. From there, they've really enhanced their inline autofill to generate, save, and fill new logins directly from the login page the moment you create them. I've noticed that. I mean, I've been using Bitwarden for years,

[02:44:39] but it's so great because when I'm on a new page, I put in my email address. It says, do you want to fill in this? Do you want to use this strong password? You don't even have to press generate anything. It does. And then it says, yeah, I'll save that for you. It couldn't be easier to create a password. It's awesome. And the opt-in fill assist feature gives you autofill accuracy for those weirdo websites. We've all seen them with unique or complex login forms. They're really focused on this.

[02:45:07] You can also very easily create an encrypted, secure export of your vault. So you've got secure backup. Business plans add things like vault health reports, integrated TOTP. You've got secure credential sharing, event logs, and admin controls. And I'll tell you for my AI, I use Bitwarden's Secrets Manager. Fantastic. It's an add-on for developer, dev, sec ops, and IT teams. If you have an AI agent that needs passwords or tokens,

[02:45:37] you know, API keys, storing them in Bitwarden's Secrets Manager is the best way to do it. And they all understand it. It makes it very simple. And because of course, very important to me, Bitwarden is open source. That means its code is available for you and anyone else to review. It's there on GitHub. You can see it all. It's also regularly audited by independent security experts. They maintain ISO 27001 certification, SOC 2 Type 2, and SOC 3 certified. They comply with GDPR, HIPAA,

[02:46:07] CCPA, so you can use it with confidence in your business. They're also big supporters of open source. They're going to host the seventh annual Open Source Security Summit. And it's this week, September 17th, and it's free. Register right now at opensourcescuritysummit.com. They've got some great speakers, New York Times White House correspondent and cybersecurity author, David Sanger. He'll talk about the geopolitics of cybersecurity. That's always fascinating. And from the Cult of the Dead Cow, one of my favorite hacker groups,

[02:46:36] author Joseph Min, he'll be discussing hackers, ethics, the security community. That's worth it right there. The price of admission. Oh, it's free. Save your seat. Just go to opensourcescuritysummit.com. Just one more way Bitwarden is committed to you. Open source and security. We love them. Bitwarden.com slash Twitter. Do me a favor. Go there. Tell your friends. Tell your family. If you have family members, I know you use a password manager. You're smart, obviously, but there are people in your family,

[02:47:06] I know there are in mine, who just say, oh, I don't want a password manager. And one of their big objections, I don't want to pay for it. Tell them, Bitwarden's free for you. It's done right. It's secure. It's easy to use. And it will protect you, please. Get started today with a free trial of a Bitwarden Teams or Enterprise plan or get started for free across all devices as an individual user. Bitwarden.com slash Twitter. That's Bitwarden.com slash Twitter. Thanksgiving's coming. That's going to be the,

[02:47:35] that's the big password discussion meal. Tell me about your password security, dad. Huh? All right. Dyson announced this, I think September 1st. It's sold out almost instantly. It is a $500 toothbrush with a camera in it. It also has a reservoir because, is this your finger? I recognize your finger. That's my finger. Do you like my very little finger?

[02:48:04] Are you using one of these? You already have it? Those are great nails. I can't get one of these. You got to review it. If you scroll down, there's a video of me. No, this was at, I went to the launch event, which was, I called an iPhone-esque launch event for a toothbrush. It was like, they had a big room, like all beautiful, like really well lit, like purple. Well, that's Dyson. They're the, the, the, the, the, the vacuum cleaner. They are, yes. And the toothbrush was, they did, they did. Did they allow you

[02:48:34] to brush your teeth with it? No, I wouldn't want to. Are you afraid of that? So the theory is the camera in the head is taking pictures really rapidly, looking for gaps. And when it sees a gap, it fires. Mouthwash. So the little reservoir. So yeah, I was able to brush on a, on a fake, you could see the fake teeth there. Yeah. And so, yeah. And it was shooting out every time you could see. Oh, that's the other thing

[02:49:04] on your phone. You can look at your teeth. You can look at your teeth with this a hundred thousand pixel resolution, which is like nothing, right? That's really, I don't want any more than that. That's fine. And that's the camera there at the top. And it has this little reservoir. And then I also, those were the fans. You also had to stand in front of the fans. They look like the sandworms. Thousand dollars. It does. It looks like a sandworm. It does. Oh my God. I know they're, they're kind of freaky looking and those things kind of open

[02:49:33] and close like the teeth. Oh, lovely. But the toothbrush shoots. So it uses something called gap optical targeting. Me too. Which analyzes 28 live images from the camera every second to find gaps between your teeth. And when, and as your, as the toothbrush passes the gaps, it shoots the mouthwash like a floss, like, it's like those, um, hydro flosser things. Yeah. Water picks. Water picks. Thank you. but see,

[02:50:02] I use a water pick. I thought, Oh, this would be great. It'll be all in one. All in one. That's the idea. Plus I'll have videos. I can send friends and family. I'm worried. Instagram. Instagram. I just don't want to think my throat is a gap. And then I'm going, it was very powerful. Like it was like, my hand kind of shook every time it was a water pick. It was sandblasting your teeth. Yeah. And it's a much, apparently a wider jet than the water pick. So it's meant to be more effective. So, um, and it's does,

[02:50:31] and it's supposed to shoot out. Um, my dentist told me to use a water wash. Yes. But, but I don't think they say a water pick is better than flossing. I think they say, no, I do both. That's true. They say, and you know, when I go in, they say, Oh, you're using both. Aren't you? I say, yes, I am. They said, my dentist said, Leo, they said, good boy, Leo. They said, he said, you're going to keep your teeth till you're dead. Oh, which I thought was a very sweet thing to say.

[02:51:00] But you have to say, great teeth you got there. Sorry. The idea is that, um, you also can see inside your mouth. So if you have any issues and you can take videos and send them to your dentist, I mean, no, my dentist loves it when I do that. It is all local. So there isn't any flock integration. Um, it is all like just how long before DHS has videos of my teeth. Let's face it. Between your phone and the toothbrush, um, and very low resolution.

[02:51:28] And then Dyson has formulated some special toothpaste. That's the, no, you need to buy Dyson toothpaste and mouthwash. And, and, and obviously the heads as well. And you, you can use other mouthwash. It's not. But if it foams, then the camera won't be able to see. The toothpaste, yes, is low foaming, but who really needs to see? You don't really need to see the camera. I can imagine that would be fun once. And then you're not going to. I had, I had a toothbrush that, uh,

[02:51:59] it didn't take pictures, but it would tell you where you were in your mouth. You know, I had this app and all this stuff. I used it once. And it does all of that. And I said, I don't need that. I just brushed my teeth. It's so, but the thing, the interesting thing is, so these did, you said they sold out, but actually, I mean, the, the theory is, um, some early people got hold of theirs. I think they sent some to influencers and also, um, a few people got them in that first week. And within days, uh, they started failing. Oh no.

[02:52:29] Because it sounds like something wasn't properly waterproofed and water was getting into it. To the battery compartment and causing it to fail within like 30 seconds. Well, so I'm glad I didn't get mine yet. I mean, I ordered it that day, but it. Well, so I think, I mean, so I reached out to Dyson and they said that there, there have been some complaints, um, that some of the early batches of Dyson new is called camera jet is the name of the product. Um,

[02:52:57] that may have experienced a component issue. So they weren't very specific, but basically water is getting into them. And a lot of them, it sounded like looking for judging by Reddit. There are a lot of people that have had, it's not even on Amazon anymore. It's gone. So one of the conspiracy theories is that they've pulled it because they've realized there's an issue. I may never get it before. Yeah. Before they sell them again. Um, and they're, yeah, so they're, they've been out of stock on the UK and the U S sites pretty much since the second day. Um,

[02:53:27] they went on sale. So we will, I would probably wait. Also, they won't let you return it if you use it. Cause it's, Oh yeah. It's a toothbrush. You know what? It's not on my Amazon list anymore. Oh, wait a minute. There it is. It's all it says is order received temporarily out of stock. Yeah. Dyson did say they are working with customers and will help. Of course they're going to replace it. Yeah.

[02:53:53] And a few people reported saying they were having issues because the support was saying, well, you can't return it because it's a hygiene product. But if it's failing right out of the box, um, they should be. I'm kind of surprised. It doesn't feel very Dyson. No. Unfortunately, you know, their new vacuums were all made like, so have things changed? Well, um, I, so I've covered the robot vacuum space and they released their first robot vacuum, not their first,

[02:54:22] their first robot vacuum mop last year. And they had been doing robot vacuums for a while. They did, they did it different from everyone else because they're Dyson. And they were actually very good at vacuuming. Very good. But they were not good at being robots. Um, they couldn't find their way around anything. Bad robot. Bad robot. And so this new one. Robot, you suck. That's good. But. Yeah. You're stuck. You need to actually move. Yeah. Um, and then, so they,

[02:54:49] they launched a lot of fanfare last IFA, a new robot, completely new design, um, looks like all the other robot vacuum mops out there. And it took a while, but they have finally admitted to us that it, they developed it in partnership with, uh, Picea, which is the company, the Chinese ODM that bought a robot. Oh. And also makes robot vacuums for shark. Um, well,

[02:55:19] they partner with, um, isn't shark like Dyson's arch rival enemy competitor. So Picea, so both, all of these companies say, you know, we partner, we use components from not necessarily, you know, so they, these are shark design, Dyson design, iRobot design, but they are, they are working with Picea, all three of the companies. And you can see the products look very similar. You can Picea that they all. You can Picea. Um, yeah.

[02:55:47] And so this was the first Dyson product that didn't have a Dyson motor in it because they actually used the Picea motor. I mean, that's like the one thing that you want to be other than the cyclone. You also want a robot vacuum to find its way around your house. And that seemed to be a problem that Dyson wasn't able to solve on its own. So they use this technology. The Chinese companies, the Chinese robot vacuum companies have definitely nailed navigation in homes. And so I, you know, I guess this was their,

[02:56:13] their way of getting into the market because their previous models had not done well. And they've now released the second generation. There were three of them that they launched at this event, um, of their robot vacuums, um, and mops. So, and those are all, they, they look very similar to the first generation. They've added the green stain detecting light, which, um, it has, uses UV to find stains. I love that from Picea. Yeah. And well, this is, this is,

[02:56:40] so it's not just looking like the green thing helps you see, right. But this actually looks for stains, um, using UV and then goes back and scrubs it. But shark does the same thing. So like that, it's kind of a shame because Dyson has always been known as very innovative. Oh yes. So here's the, this was the real excitement from Eva. This is what, this is a vacuum on a vacuum. It's like the vacuum with two backs. What's I, what, what's happening?

[02:57:09] This was the duo that I cared about, Leo. This was the duo for me. This was the Roomba duo. Ladies and gentlemen. So, okay. So it does look a little like R2D2. What's going on here? It does. So, um, it births, uh, a baby robot. Yeah. here it comes. Here it goes. So it is, it's for those listening.

[02:57:34] It is a larger robot vacuum that has a smaller one on top and it uses a mechanical arm to lower the smaller vacuum, um, to then send that off. So the idea is you have larger spaces in your home that need more deeper cleaning. And most robot vacuums aren't powerful enough to really clean deeply. This is like a steam clean. You've probably seen when you, like if you're in Vegas that you'll see these, um, robots roaming around, like cleaning the floors,

[02:58:04] like floor washing things. Yeah. They used to have people attached, but they've stopped. Now it's just robots. So this is kind of like an industrial strength cleaner. Um, it's a floor washer steamer and a robot. Okay. So it's not for consumers. No, it is for consumers. Um, but it's in carpets, much more carpeting. You need. So it actually, in the video, you can see it sucks up, um, some pasta. So it actually like it'll take, um, it's not just, I suck up pasta. What does it need? Uh,

[02:58:34] what do I need a vacuum for? That's, that's what I've got. I don't want the vacuum competing with me. It's too big to get around a lot of places in the home. So then when it gets to an area, like around a table, it spits out the baby vacuum. Does it pick up the baby vacuum after it's done? That's a good question. The back, baby vacuum goes back into the robot arm and then it lifts it. That seems destined to fail. It's vastly over-engineered. It is a concept.

[02:59:03] It is not actually for sale. They were waiting. It's exciting to see what reaction would be. Um, I think the, the good thing here is I, I like the main vacuum. I'm like, actually that as a, there is, it's very similar to the Matic, which actually is one of my favorite. I was going to say that one's getting a lot. It has the big wheels so it can actually get around and it is more powerful. Cause one of the problems with today's robot vacuums is they, they talk about their suction power and stuff,

[02:59:32] but they're actually not as good as just using a stick vac. That's why you, but if you run them every day, it will keep your house cleaner. But this one is designed to really tackle, you know, tackle more dirt and grime in your home and bigger messes. It's a wet dry vac. So that part I think is quite interesting. I, you don't, I, you don't need the little baby one on top. I mean, that just seems a little, I kind of want one that has like maybe six DVD sized robots that it sends out. And they all just go around and clean. Little flying ones. Yeah.

[03:00:02] Drone, drone robot vacuums that can just go up the stairs. You know, so, so this isn't Roomba, the Roomba we know. This is a Roomba that went bankrupt and was bought by a Chinese company, basically. Yes, correct. So there, so Roomba does still. They were put out of business by Chinese vacuums. They were, they were, so they went bankrupt basically. Well, iRobot is the company and iRobot was tried to be purchased by Amazon. And then that fell through.

[03:00:33] This was about three, three years ago now, because of government in the, was it the FCC? One of the government organizations. I think the FTC. FTC. Thank you. Getting my F's confused. The FTC blocked the merger, as did some of the European equivalent. They said it would be too likely to create a monopoly for Amazon. And that basically was the nail in the coffin. Because at this point, iRobot had been declining its market share.

[03:01:02] It used to be the dominant robot vacuum company. And they invented the category, didn't they? I mean, they were the first. Very much. I mean, there was an Electrolux one that came out like a few weeks before, but yeah, the Roomba was the main, the main one. So yeah, they went bankrupt at the end of last year. And they were bought by their major supplier, which was Picea. But the company, iRobot, still exists. They still have their Boston headquarters, but the majority of their development and products,

[03:01:32] and all the people at this booth were all Picea engineers. So yeah, it's pretty much been, they have a new CEO who's from Picea. So it's been very much taken over by. Can you even buy these? Didn't they block? Doesn't the F, doesn't the federal government? So yes, this is the other thing that's happened to robot vacuums and robot lawnmowers and all robots. If you were hoping to have the humanoid in your home next year, that everyone's promising us to do your housework, the FTC has banned. No,

[03:02:01] now this is the FCC. Sorry. No wonder you get it confused. The FCC has banned all mobile robots manufactured overseas. Does that include robot vacuums? Over a certain weight. Yes, it includes robot vacuums and robot lawnmowers. You most, so I was at EFA this last week and several of the robot vacuum manufacturers were there. And they said the majority of the products they're releasing right now have already cleared FCC

[03:02:29] because they went through before the ruling came down, which was in July. This is, they've been added to the, the, what is the list? It has a name, the list that, this is the same for the drones. The covered, covered list. Thank you, well, I mean,

[03:02:58] if you are concerned about privacy, a robot vacuum, privacy and security. They're not concerned about privacy. Look at the clock cameras. Look at the clear view AI. This is what they said. They don't care about, look at data brokers. They don't care about privacy. So the FCC did say, you know, privacy and security, these mobile robots have cameras and, you know, but none of the ruling actually prevents.

[03:03:27] And there's nothing in it that makes the company have to secure or do better privacy. You just have to ask. You just have to promise to start making your products in the U S that's essentially a someday, but that that's essentially what they're asking. And so it's more, it's more of the sort of the economic side rather than the privacy and security. And this is why I don't want the federal government to nationalize AI because if they were halfway smart, okay,

[03:03:55] but they're just greedy sons of bitches and they're just stupid. Well, didn't. So one of the rats, a manufacturer. So was it TP link? He got, he paid, he bought the CEO of TP link did the golden CEO ticket thing. Yeah. And then they got there. Yeah. They got approved. Look at that. Oh, would have thought it. I think they did. How much was it? $5 million. It's a little bribe is all it takes. And then a Zeus just got it this week. And then the very first robot got it.

[03:04:25] Husqvarna, the Husqvarna law, which is it by the way, that's a Chinese company, right? So they're originally Swedish. Swedish. Oh, the Swedes are hacking us. We're in deep, deep trouble, but they do have rather large American operations. Actually they're in North Carolina. So there's Swedish from North Carolina, but they're hacking us. How much do they pay to get approved? So, yeah. And they didn't,

[03:04:52] the interesting thing about the robots is there's no limit. So Husqvarna now has just, they're fine with, for all of their products going forward. Whereas the routers have a limit. I think it's 2028. I think your, your coverage expires. But like everything else, they're made in China. So that's the issue, right? Everything's made in China. Everything's made in China. Everything's made in China. And, but, and so I think what, when you fill out these forms, you have to promise to try and start building your products and,

[03:05:22] and show some kind of roadmap for building products in the U S. But it doesn't sound like anyone's really done that. Well, what's hysterical is in this press release from the FCC. It says we have a determination by the department of war that these devices do not pose risks to national security. I mean, if there's any robot that proposes any risk, I would say the one with the camera and the blades probably. Yeah. Yeah, that's true. We could be the attack of the killer lawnmowers. So it,

[03:05:52] the whole thing is kind of a farce. I spoke to a number of robot vacuum manufacturers at EFA and they were all just like, we're just going to wait and see because we just, we think this is all going to change at some point. I think they were quite, the Chinese ones in particular were thinking hopefully that the, cause there was a summit between Trump and the Chinese. Yeah. And I think they were hoping that something might come out of that to address this, because it doesn't all feel like politics rather than any actual consumer. It's totally politics.

[03:06:21] They're not protecting us. They don't care. And you know, taking away people's robot vacuums is going to make them mad. Cause people. exactly. You're coming into my home and taking away my robot vacuums. I've got five of them right now. I bet they've done research. They will not take, they will not come and take. They're not going to come and take it. First they took the robot vacuums. I forgot. And I didn't care because I wasn't a robot. The flock doesn't understand sarcasm. But yeah, the, the,

[03:06:49] the Roomba duo may not be doing into your life anytime soon, because it is both not a real product yet and would not be covered under the current exemption. So they would have to get, they would have to get an exemption from the covered list in order to get this one into your home. So it's going to be interesting. So you have to send one to Pete Hegseth so he can verify that it is not a national security threat. Then we can get it. Maybe that's what it was. There's some, there's some leader out there. You know, the Hegseths have the cleanest house.

[03:07:19] Yeah, exactly. So I've got a mansion. I need one in every room. There are so many. I saw so many robots at Aoife this year, which was weird. You don't normally see a lot of robots there, but the humanoid thing was, it was definitely, I mean, did you see a lot of bipedal robots? Yeah. Lots. They had a robot runway. They had them like do going down the runway, doing back tackling people. Do you think this is a category really? Or is this like going to be like nerd helmets? It's just going to be something that companies thought we wanted. And nobody really wants that.

[03:07:49] Nobody wants a giant, super strong humanoid in their house. No, I actually wrote a piece last week before Aoife saying the headline was the robot butler revolution does not have legs. I wouldn't be, I wouldn't be unhappy if it didn't have legs. So I don't, I really don't think that was my point. At least you have an escape route.

[03:08:17] I think home robots are going to take off. I think we're going to get, but they won't be bipedal. I don't think the full on humanoid is really ever going to happen. And they're too complicated. They're too expensive. They don't do individual jobs. Well, yet every single one I saw was remote controlled by a PlayStation controller. So, you know, they're not, you have to train them individually. Oh, that's an improvement. I'm not going to do the laundry myself. I'm going to use a PlayStation controller. I'll just use a PlayStation controller to control the robot to do it for me.

[03:08:47] And every time I see these robots doing laundry, filling dishwashers, folding clothes, they do it painfully slowly. very slowly. It's like you want to get out of the way. Just let me do it. Get out of here. This is the problem. So this is what the company say is we can train these robots to do the specific tasks. There was even one booth where they gave me like the, the controllers and the headset. And the idea is, you know, you go and do the task, do it a thousand times. And then the robots learned how to do it. I'm like, this isn't a plug and play consumer robot device.

[03:09:16] If I'm going to spend $3,000 on a robot, that's going to do my laundry. I want to come, I want to just press a button and it goes and does it. I don't want to have to spend three weeks training it. And so I think I'll get a dog. If I need to train something, I think robots that are going to be able to do specific tasks. Well, we're going to start to see more of, I mean, we've seen the laundry folding robot at CES for like 10 years. Eventually we will get to one that actually works. They will get there eventually.

[03:09:44] Dreamy has a robot that can dreamy is the robot vacuum manufacturer. And they have one they showed off that can, that can move the laundry from the dish, from the dryer into the, from the washing machine into the dryer. Oh, thank God. I hate doing that. I hate doing that. That task. So I think we'll start to see more of those. LG has its cloyed robot, which actually uses its, it connects to all the smart appliances in your home and communicates with them

[03:10:13] and sort of gives them commands. So it's more like a smart home hub, but it can also move around, but it's on wheels, not legs. I want Cloyd, the robot barber that goes, okay. I'm going to come and give you a haircut. Yeah, no thanks. So people, it's like AI though. People, well, it is AI because this is why we're seeing so much development here is because the world models that they're able to start to train these robots on. We're seeing, I mean, physical AI,

[03:10:42] this is the manifestation, but so there's people are throwing an awful lot of money at, at these companies. Whether anyone's actually going to come out with something that a consumer level is going to be useful. I'm still very skeptical of. No, me too. I'm glad to hear you say that. I mean, I love AI as you know, I'm kind of all into AI, but this, the bipedal robots, nobody wants that in their house. It's terrifying. Yeah. I,

[03:11:09] having gotten up close and personal with some of these robots, I definitely agree. Yeah. You've been tackled by, I could see it in an industrial environment. Yeah. And I know when I'm in the nursing home, they're probably going to be caring for me. They'll be helping you stand up. Yeah. I imagine 20 years from now, I'll be going, hello robot. But. Well, the companion robot thing is the big, big push we're seeing. A lot of, you know,

[03:11:34] the robot that will communicate with the senior living alone, sort of give them companionship, help keep their mind. I can see that. There's a lot of that. And we've talked about LEQ on the show before. That's one that has done, done really well in this space. And there was one from Tuya at Aoife that was a companion robot that can actually move around and follow you and see if you fall. That's funny. Cause that's how I thought you pronounced your name for years. Tuya. Yeah. No,

[03:12:03] I have devices somewhere in my house. I have a Tuya device, but I don't have any Tuya labeled stuff. Some company has, something is using Tuya, but I don't know what it is. Tuya is like a, they're a big cloud platform. And they, yeah, I mean, but you don't actually buy that. It's a light switch or it's something. And I'm not sure. I don't know. I see it on my home assistant, but I, you know, I push the button and nothing happens. I don't know what it's. You probably bought it off AliExpress, didn't you? Cause it's, yeah.

[03:12:33] Or the guy who built the house did. Yeah. Hey, I feel bad. This has gone on way too long, but it's so much fun. You got me going on robots. Sorry. It's so much fun talking, not just you. I mean the whole thing. I appreciate your very dear patience. This is a world record beating show. And we've practically lost Daniel. I don't know if he's still alive. It's actually frozen. That's not. He's secretly having dinner and he's put his picture up. And no, it's so nice of you to be here. Daniel Rubino from windows central.

[03:13:03] Thank you so much. We really appreciate it. He's editor in chief there. Your insights are always welcome. And I knew that you would have some smart things to say about this. AI doom that we are all facing. Thank you, Daniel. I appreciate it. Thank you, Mike. I'm thrilled to have you as a partner at a twit. And of course, can't wait to see a hands-on AI October 1st. What's, do you know what your first show is going to be? Yeah. It is going to be a basic understanding of LLMs. We got to, we got to lay a foundation. How these things work.

[03:13:33] Yeah. How they work. Yeah. I don't think it's widely known. And we just assume, well, you put something in and something comes out, but I think there's more to it. I suspect I may be wrong. It's a black box. There's gotta be something in there. Thank you, Marka. Appreciate it. See you soon. And of course, Jennifer Patterson, Tui and her robots and her.

[03:14:07] That's the reason why they will not take over the world because they'll. Because you have to change the batteries. That's true. They don't have nuclear batteries yet. Yeah. Although I'm excited to see the Clara and the Sun movie that's coming out. I love the book. The book was wonderful, but I love all of, was it Isakuro? Isakuro's books are just right. He wrote the remains of the day as well. Yes. And Clara and the Sun was just wonderful. It's about a robot. Mm-hmm. A companion robot. A companion robot that uses the sun. Hence. Yeah.

[03:14:36] Thank you, Jennifer. Thank you, Daniel. It was a lot of fun. Micah. Thanks to all of you for being here. Special thanks to our Club Twit members who make all of this possible. If you're not a member of Club Twit, please consider joining. Without you, there would be no Twit. 10 bucks a month gets you ad-free versions of everything we do. Special programming in our Club Twit. The club is a Discord server. It's really a fun place to hang out, but also that's where Micah and I, for instance, did the Apple event. If you are not a member,

[03:15:06] I urge you to visit twit.tv slash Club Twit. If you listen to our shows and you want to support what we do, you want us to keep doing it, it's going to take your help. Twit.tv slash Club Twit. And thanks to all the members. We appreciate your support. We do Twit every Sunday afternoon, 2 to, well, 5.30 in the afternoon Pacific time. That's 5 to 8.30 East Coast time. It's bedtime, practically, for you East Coasters. We also do it, that's what, 2100 UTC.

[03:15:35] You figure out what your time zone is. You can watch us live on YouTube, twitchx.com, Facebook, LinkedIn, Kick, or if you're in the club, in the club, Twit Discord. After the fact, on-demand versions of the show are available at our website, twit.tv. There's a YouTube channel with a video. And of course, it's the best way to get any of our shows. Subscribe in your favorite podcast player. Let's keep podcasting free and RSS based, and that's the best way to do it. No spying on you,

[03:16:03] just content at your convenience. So please subscribe. We appreciate it. Thanks for joining us, everybody. We'll see you next time. Another twit is in the can. bye. This is amazing. Bye. Bye.