Anthropic is Introducing Text Watermarking… But How? - DTNS 5329

This is the daily tech news for Tuesday, August 11th, [music] 2026. We tell you what you need to know, give you the important context, and help each other understand. >> Yeah, here's something I want to understand. Today, Anthropic announced that it's rolling out text watermarking. >> Uh huh. >> How do you how do you watermark a text without changing it? And they're also doing C2PA. Um >> that's a real good question. EU made him do it, but all of us get to benefit. >> Mhm. We're going to talk all about it coming up here. I'm Jason Howell. >> I'm Tom Merritt. >> Let's start with what you need to know with the big story. >> So yes, Tom, as you alluded to uh Anthropic shared in a help document uh that its models will carry embedded text watermarks also C2PA signed file metadata. this is all in compliance with the EU AI act. But that might make you think okay so they're going to get it the rest of the country the rest of the world will not. Uh but that is not true. Apparently the watermarking going to be applied globally for all users. So this is one of those cases where that actually happens. Uh the watermarking covers the cloud API cloud cloud code cloud co-work tag third-party host. So, if you get your cloud through something like AWS or Google Cloud, um it's going to survive through there as well, I guess, because it's it's text. Yeah, I I don't know how this is going to work, but Anthropic says that the text mark survives copy paste actions uh and some editing without changing the meaning or changing the quality of the text that it's attached to. It's man, it's so puzzling. Well, Anthropic also says it'll support its users in detecting Claude's own watermarks with details on how that's done coming sometime in the future. So, we don't even know quite yet how you would even detect for this, but Anthropic's [snorts] going to have a tool for you. Anthropic says the watermarking will come to new models, of course, but also they're working on integrating it into earlier models that were released prior to this August 2nd um deadline that was required by the AI act. So yeah, text watermarking I I mean I'm sure text watermarking is a thing before this, right? >> It is a thing. Um, in fact, Apple mo more famously than most used it to watermark some of its internal documents which is how they were able to track a leak >> because they could find the text watermark. Uh, as I understand it and and I you know I I haven't spent my entire life studying it, but [laughter] as I understand it, text watermarking takes a few different forms. It usually has something to do with word frequency or patterns. Uh so in in the English language, we have so many synonyms. I imagine it might be a little easier, but you can use synonyms or use certain phrases and then you can you can keep the meaning the same while changing the words to make sure that later you can do a text analysis and go ah that was made by Claude that because that's Claude's way of using those words in that that manner. uh similar to how you can actually identify the writer of something based on their word choice and their word usage, right? Watermarking is kind of like that. >> Um >> I guess that doesn't change the meaning, but it actually as someone who wants to use these tools for like just grammar, you know, and word, I don't want it changing things to watermark them. So like yes, just make things bold, but don't change any of my words. Uh, and if it's trying to watermark, then it might have to change words. So, I I I guess I'm not trying to be like, and therefore I'll never use Claude, but I'm like, I need to know a little more about how their kind of text watermarking works, >> right? And and when is it applied? Is it like, here's my paragraph. Uh, make sure that I didn't miss any commas. And it inserts a comma. It's like, all right, text watermark in there. We had something to do with this. >> Yeah. Yeah. I I'm guessing it doesn't work like that. There are some other text watermarking that use invisible characters. Um, you know, that's what I was wondering. That's when they say like it'll survive most cut and paste. I'm like, "Oh, okay. So, maybe they're just maybe they're just putting, you know, unreadable characters in there and then when you copy paste them, they that persists." I wouldn't mind that as much because it's not messing with my copy then. Um, but yeah, I I'm I'm very curious how the text watermarking works. I know that text watermarking works. I just like which one are you using? It makes a difference to me >> as you're describing it. What comes to mind for me is I feel like AI already has a watermark. It's all of the things that we like. >> It used to be the M dash, right? Yeah. >> Yeah. It used to be the M dash, which you know, of course, people who love the Mdash and have used it forever would push back on that, but um speaking of that, you are right to push back on that, Tom. You know, like [clears throat] things like that feel like, you know, the the AI output >> watermark [laughter] and up themselves. Um, but yeah, I'll be really curious to see like what the impact of something like this has on usage patterns. We're we are kind of in this moment where it's like we got your frontier models that everybody's used to using in a short short super short period of time in the cloud, super powerful. We use them for everything and then now we have this growing and we'll talk about it, you know, certainly throughout this show. Um, this kind of the percolating open model uh, you know, localized model thing that's happening. Does this do do these watermark efforts end up pushing more people inevitably towards those more open platforms where that doesn't exist? Because some people are going to be very resistant towards watermarking even if it's a good like good idea in practice or in theory to have them. You know, >> the the answer is always a little bit, right? The question is always how much. It will not push everyone in the world off of Claude, >> right? I say that uh with with full confidence. Uh it will push some people off of Claude to use an openw weight model that isn't subject to the EU AI act. Uh I say that with full confidence. The the the amount I imagine it's going to be a small percentage, but we talking like a 0.01 or 1% like that that could make a difference as well. And I'm curious how Open AI and Gemini and uh whatever Meta is going to call their model today. Uh I I'm I'm curious how they are going to comply with this as well. >> Almost certainly we're going to see more of this. But yeah, I I am very curious to see, you know, what that watermarking actually is. Is it the invisible text? Is it, you know, some sort of phrasing or or patterning? I hadn't even considered the patterning thing, but all of that all of that makes an imprint in one way or another on what is being read on the other side. So I Yeah, >> there's always there's always the criticisms of like, yeah, and you can rip it off, right? You can you can remove the water marking, which is true of files and and images as well. Uh, and it's like, yeah, okay, now we're getting to whether the EU AI act is actually effective or not. Uh, and and usually these things are usually attacked as like they won't work all the time and then they are defended as they're not supposed to work all the time. They're just a speed bump. uh you know and and so yeah I I don't know that these things are impossible to remove but they they do help to trace things. It's just they're not perfect and and they never will be. I don't think >> Yeah. Yeah. I'd agree >> that I'm less confident. Maybe someday they will be perfect but I don't think they will. >> Maybe. Maybe. >> You know who is perfect though, Jason? >> I do. >> Danny Verer. >> Oh, I was going to say you, Tom, but >> Oh, no. No. You're right. You're right, Danny. >> I'm close to perfect. And the thing that's getting me close to perfect is Howard Yish. >> Yeah, that's true. Um, but don't don't don't leave out John Atwood cuz >> No, we wouldn't perfect >> without John Atwood or Pat. >> Or Pat. No, you were like 99% of the way there. Pat took you to 100. >> Yeah. Thank you, Danny, Howard, John, Pat, and every single person who supports us in all the various ways you can, including patreon.com/dtns. >> All right, so much more we need to know today without watermark. Let's uh get to the briefs. Yes, Windows latest reporting uh on a Lenovo patent filed back in August 2024 granted in March 2026 that points to a think book cenamed project swan. >> It's a think book that will float elegantly on [laughter] on a lake. Uh no, actually uh it it actually expands its display sideways like swan wings maybe. >> Okay. >> Yeah, sure. You press a button and then the you open your laptop, press a button and then the screen expands. Uh that extra display would be housed on both sides of the lid with motorized rails extending them outward on demand. Uh has Thinkbook label. So it puts this laptop in the business and productivity lineup. Uh cuz Lenovo already has a version of this that for gaming. Uh in fact, they already sell the Thinkbook Plus Gen 6 rollable. That one extends vertically. So that's that's a little different. Uh and it adds around 50% more screen area when it's unrolled. Um so you can do a rollable screen in the market. Lenovo also showed a horizontal display like we're talking about at CES 2026. Uh calling it a proof of concept at the time. The Legion Pro rollable had a 16-in OLED expanded sideways to 21.5 in and then with a little boost could go all the way to 24 in. Then there's the Thinkbook Flip. That is a concept that used a single OLED panel that bent over the top of the lid. Uh there's the Legion Go Fold that they demonstrated, a 7.7 in P OLED panel that unfolded to 11.6 in. Anyway, getting back to this one, Windows Latest talked about Project Swan. We don't have a panel size. We don't have hardware details. There's no price or anything at this point. Windows Latest just thinks that not only is this patent going to turn into a product, but that product is close to an official announcement. Yeah. And when I think of um when I think of certain certain reasons that I use computers, you know, in the like production space, I think of like timelines, like video timelines, and you know, like the vertical extension great for I think coders. Like I'm not a coder, but that's that's always the example that I that I feel like I see. But that doesn't really apply to me as much. I've always been like h why would I ever need a vertical horizontal though, man? Like I would love that because like I've had the the like 17in, you know, MacBook Pro. It's just a massive behemoth in my bag, you know? I would love a smaller uh more compact laptop by comparison that could extend out um on a horizontal sense because, you know, like timeline editing and that sort of like horizontal uh workspace in certain applications is just incredibly useful to have. I love this. Yeah, I I when I am on the road and using my laptop, which is a 16inch screen, uh I am still thinking gosh, there are some things that are easier with my big wide Dell monitor at home. And so it would be nice to have the option where like yes, I can use the laptop in a confined space like a train or an airplane and and not have to deal with a big wide screen like the 17 in, right? Um, but then when I want it, like I'm sitting at a desk in a hotel room, I could then press the button and open it up into that wider screen. I think that would be good. The question is like Lenovo seems to be the one that's really pioneering this uh and really pushing it. And they're still only got one of these models out. Maybe this is the year we start to see more. I assume we're going to see more at CES 2027. Uh, but it's taking a while for this to come from phones where we we've kind of recently said like, yeah, okay, foldable, that's that's a legit product category for phones these days. Uh, and and it hasn't quite made it to the desktop space or the laptop space. >> Yeah. And durability is also a thing, you know, that's a that's an area where >> that was a thing with the phones, too, right? You had to prove that it wasn't going to just break on you. >> Yeah. You take it to the wrong place and you get a windstorm with a sandy area and suddenly those gears that [laughter] It really is swan. >> Yes. >> It goes flying. >> It's good. Yeah. All All you can get out of it anymore is just chuck it in the water and watch it float. >> Yeah. Yeah. >> Uh anyways, I don't know. But I do think the concept is really cool. Um so I love the idea of that. >> Yeah. And I'm I'm glad Lenovo is pushing the barriers on this. >> Indeed. Yep. >> Or boundaries. Boundaries. I think I meant boundaries, not barriers. >> Yeah. Yeah. It worked for me. It worked for me. Apple's iOS 27 uh beta 5 release yesterday included hidden code that referenced six upcoming iPhone models buried inside a battery driver and battery intelligence system files. I'm sure you would never imagine what those models were. iPhone Air 2, iPhone 18 Pro, iPhone 18 Pro Max, iPhone 18, I know, uh iPhone Ultra, the foldable, and and uh the iPhone 18e. So, okay. See, those those names, it's not like they're new names and we hadn't heard them or expected them. But the leak uh does signal uh separately from this a split launch approach, which basically means that the premium models would be landing this fall as you know, at the time that we normally expect to see them, while the cheaper ones, including the iPhone 18 and the 18e would be pushed until early 2027. That's a little interesting. Um, Bloomberg's Mark German also reported that Apple will release a glass ccentric iPhone to mark the 20th anniversary of the iPhone, even in light of production challenges that have made it apparently difficult to execute. Jeffre had a report that Apple had backed away from an all glass version. Um the road map at least according to Gurin is still unch unchanged though we did acknowledge that a version with even more glass had to be abandoned because it couldn't or wouldn't work at scale. So >> yeah go read the 9to5 Mac story uh analyzing the German and Jeff disagreement and showing that it's not a disagreement. This is a uh this is a clash in headlines. Uh Jeffre is saying there was an a actually all glass version that Apple has abandoned. And Gurman's like yes there was an all glass version that Apple abandoned because they couldn't make it work at scale. Uh but they are still going to do a mostly he's only mostly glass uh version next year. So the this this is how this stuff works and this is if you're not paying attention you get confused by seeing all these conflicting headlines when they're not really conflicting. Uh so I I think I really appreciate it. 9 to5 Mac kind of breaking it down that way. The other stuff is like like you said it's like oh my god we're shocked uh Apple's going to announce iPhones in September. >> No way. >> As they have since I was a baby. >> Yes. >> What about the split release? What do you think about that? >> The split release is something Gurin's been reporting on for for a while now. So I'm not shocked to see that continued to be reported on. But that is at least a significant change, right, of like, oh, we're going to move the iPhone 18 off into the cheap area. uh the SE and now the E and the Air have traditionally been in the spring. So, it's not shocking that those would stay there. Uh but moving the base model iPhone there is an interesting move if that's what happens. Uh I think that will be interesting to see. So, yeah, it's just it it's funny. We've got a Pixel announcement coming tomorrow afternoon. So, we we won't talk about it on the show until Thursday. Uh, and in advance of that announcement and because it's August, I feel like all the news just [laughter] is like holding back like like like and so we end up getting these like Apple's going to release iPhones like >> no way. >> Fancy. >> Yeah. >> But the glass thing I think is interesting. I think it it would be interesting to see that. >> Now, if they could do an all glass foldable, that would be even more interesting. >> Now you're getting ahead of yourself. How do you do a glass hinge that doesn't make people cringe? >> I don't know. >> I didn't mean for that. I just couldn't. Yeah. [laughter] >> Uh let's talk about OpenAI. Uh you may not realize OpenAI has its own security preview program called Daybreak. This is basically OpenAI's version of Anthropics Project Glasswing. Uh Daybreak has been split into two tiers. Now Daybreak Blue is for defensive work. So, finding vulnerabilities, reviewing code, uh, and the folks in the Daybreak Blue program, uh, get access to GPT 5.6 SUL, uh, so we've actually mentioned that that was released to this limited group, uh, in the past, but has not been released generally. Daybreak Red is for actually like finding new vulnerabilities. This is not defensive work. This is kind of proactive work. Uh, it's for exploit validation. It's it's for it's for finding problems. And there is a new model coming to the Daybreak Red folks called GPT 5.6 cyber which is built on 5.6 solve but tuned for those kind of security tasks and will not refuse to do security tasks. Uh there was there was a report that even GPT 5.6 saw for Daybreak folks refused to do security tasks like 98% of the time. GPT 5.6 six cyber will not it does your security test 95% of the time uh so hence the reason that it is limited to the daybreak folks and there are more daybreak folks now uh so uh along if you're in Daybreak you get to do blue or red you just pick it but Accenture IBM crowdstrike cisco sofos and cloudflare are all now in the daybreak program and can use the models to protect their services and customers you can actually use daybreak too if you are a customer of one of the companies in Daybreak. You just go to them and say, "I would like to submit this project for you to use in Daybreak and then they do it for you." Uh, OpenAI says, "Safeguards on Daybreak Red, so on the cyber version, include identity verification, clearly defined testing, scopes, logging, monitoring, and human oversight. Uh, so they're not just letting it run wild as we have seen other agents do in the past. >> Why not? Come on. There's nothing to be worried about. What could possibly go wrong here? How much of this stuff is um like the and then this is total like guessing but you know we've got the hugging face we've got these what you were literally just talking about agents kind of breaking out of sandbox blah blah blah this has been such a story over the last couple of weeks with open AI like how much how much of these these news items are tied to that in a way of of you know for open AI to say hey look we're responding we're doing the right thing or are these things just already in the works and the timing is, you know what I mean? It just just happens to be lined up because these things do take a while to get there. But I saw some criticism of like, oh, they're just announcing this right now because of the whole hugging face thing and they want to, you know, change change people's minds on their responsibility, the company, but I'm sure they've been working on this for a long time. >> Yeah. I'm sure the person you saw say that knows way more than anyone. >> Oh, yeah. Totally. [laughter] >> Uh, >> but but I think it's representative, you know, >> it's not like we live in a world where everyone can have an opinion on anything without any backing of it. >> Yeah. Totally. Yeah, I I think it's both. I think you don't have a model like cyber that you just whip up since Friday, right? Yeah, totally. This is something long in the works. >> And I agree. >> Yeah, you can massage release and announce dates for sure, but you also have clientele. You have these people in Project Daybreak and you've probably been briefing them uh confidentially about these capabilities coming and you want to deliver on that. So, it's it's not like you can just do whatever you want. Uh Yeah, you could have announced it maybe Tuesday versus Thursday or whatever, but there's a limit to the window in which you would announce it. So, no, I don't think they only are announcing this because they want to get ahead of the news. However, a lot of the way they announced it probably does have to do with that. Like for instance, announcing these safeguards and saying, "Look, it's really capable. It won't refuse to do as many things because we know that was a problem Hugging Face had, right? So, let's let's make sure we know that like, hey, this one will work better, but also [laughter] uh we have safeguards and there's identity verification and human oversight, so if it does start to go rogue, we'll see it. Uh I I think they're they're definitely tuning their message to react to the situation that's out there. >> Yep. And they they all have this uh this kind of cyber um arm. Well, maybe not all of them, but it's certainly becoming um very very quickly this this like cyber part of the business, the model that is so hyper like capable of of detecting and and doing all this work. I guess what I'm saying, I'm having a horrible time saying it, is I'm I'm impressed by how quickly the industry moves to these different directions. And it really feels like in the last couple of months suddenly this has become a very big part of their focus and their business on on all the frontier models. You know >> I think security vulnerability finding obviously anthropic started the drum beat with fable uh and mythos right uh and I think that is something that people are very excited and concerned about like I'm excited to use a tool to tighten up the security in my network. I am scared that a bad guy [laughter] is going to use the same tool to attack my network. Right. So there's there's a lot of focus on that and I think that that it's natural for the industry to move that way. But you're right. It just like everything else in this space. Uh things things are moving at a really really quick pace. >> Yeah, indeed. >> Uh we have an AI LLM channel in the DTNS Discord uh that has been hopping lately. In fact, uh, [laughter] over the past day or so, it's taken a turn into is the world worse now than it's ever been kind of conversation. And of course, as you might guess, if you know me at all, I'm the guy going like, it was pretty bad in the past. Like, yes, there are bad things, but maybe it's actually not as bad as it has been in the past. Uh, and of course, I'm a Gen Xer, so I'm talking about like, hey, when I was a teen, I thought we'd all die in a nuclear war before I got to be an adult. So, you know, some things have improved. Anyway, I say that to encourage you to join our Discord. Uh there's a lot of really good conversations happening there and you can join it by linking a Patreon account at patreon.com/dtns. All right, we have a bunch of quick headlines that are good to know and will guaranteed make you look smarter. >> I don't know if we can guarantee Well, I make you look smarter. >> I will personally guarantee that you will look smarter. Will you be smarter? [laughter] Okay. All right. I'm with you. >> Uh, Google is testing a version of its homepage for some signed out users. There's a very few. It's actually kind of hard to replicate, but if you sign out, you might run into a test version of the Google page that has no search button. Instead, you get four buttons. Create images, ask about files, brainstorm, and the classic I am feeling lucky. >> Interesting. Yeah, >> that's so fascinating to me. >> And they're testing it, so they're trying to see what effect it has. But yeah, it's fascinating. >> Yeah, Bumble changed the rule that differentiated it at its launch years ago. Now, anyone can send the first message if the app matches you to someone. This is a dating app, of course. Uh, since its launch in 2014, only women could send the first message. Bumble says 66% of women surveyed said they prefer men to send the first message. >> It's not even Bumble anymore then. >> Yeah. What is it? It's like >> I don't know. I haven't dated since like Clinton was president. So, >> [laughter] >> um, His Majesty's Courts and Tribunals Service, which runs civil and family courts in the UK, has banned meta smart glasses from court buildings across England and Wales. kind of interesting because you can bring a phone into the court. You just if you as long as you're not using it for recording because recording is not allowed. Uh but you can't even bring the glasses in cuz they're like we we it's too hard to tell if you started recording or not which it's got the light but whatever. >> Yeah. The phone's in your in your pocket. The glasses with the camera is just always on your face. It's it's hard to ignore. >> Even if the phone's not in your pocket, [snorts] >> it's it's certainly easy to see if you're trying to do video. And it's probably easy to like walk around and look at the screen. So, >> it's easier to pretend that nothing's happening with a phone when it's a phone versus glasses, >> right? [snorts] >> Yeah. Uh, Anthropic says an unreleased model made significant progress towards a solution to the Ryman hypothesis, improving a mathematical lower bound on the proportion of zeros in the critical line for the Ryman Zeta function from 41.6% to 67.2%. Anthropic uh released a paper actually showing its work. If none of that made sense, an unreleased claude just made progress on that unsolvable problem from the 2017 movie Gifted. So, THERE YOU GO. MADE PROGRESS. >> OH, OKAY. That's kind of a big deal. Also, especially because LLMs are not naturally that good at math. >> Yeah. But there's more and more of these stories happening where it's like, oh, how do mathematicians >> I guess I can't say that anymore. They they're proving me wrong. >> No. Yeah. >> Uh just like my friend in high school. Uh Samsung says the US FDA has cleared a hearing aid feature for the Galaxy Buds. Uh that means that the Galaxy Buds 3 Pro and the Galaxy Buds 4 Pro will get hearing aid features in some markets later this year. >> Love it. Love it. The US 9th Circuit Court of Appeals ruled that section 230 of the Communications Decency Act is quote a defense against liability, not blanket immunity from being sued. End quote. Thousands of lawsuits against Meta, Tik Tok, Snap, and YouTube alleging youth addiction may now proceed as a result. >> Yeah. And this doesn't mean they couldn't use uh Safe Harbor as a defense. They just have to use it as a defense in the actual lawsuit, not not as a way to stop the lawsuit from happening. >> Uh, back in September or back in September, [laughter] I'm a time traveler, apparently. Back in September 2026, for those of you watching, probably in the future, Spotify will roll out AI persona badges for artists who disclose that their public identity is generated. And Spotify said if they discover that your public identity is generated, they might give you the AI persona badge as well. Uh this goes along with the verified badge for a human artist that can prove they're human and badges that identify generated music whether the artist is human or not. A system called Signal Trace from a company called Leonardo is being marketed as a way to collect signals from Bluetooth and other transmissions uh to create an anonymous way to identify movement. Now, paired with things like license plate readers, it could uncover the identity of the person who's using the devices. >> This is similar to using Bluetooth to track you around a mall to serve advertising. And it's the same concern about like, yes, your Bluetooth identifier is anonymous. It doesn't carry your ID, but they could pair it with your credit card transaction at Zara, and then suddenly true >> they know who you are. Uh, finally, the $99 MagSafe slide out keyboard attachment for your phone called Clicks Power Keyboard, the one we saw at CES, has now shipped 75% of pre-orders. >> Oh, right on. It's real. >> I love I love a CES product that shipped story. >> Yes. And uh and Clicks products are are pretty cool. I really wish them well. >> Uh we end every episode of DTNS with some shared perspectives. Today, Brian H has a thought about smart glasses getting banned from certain areas. What we were just talking about a few minutes ago. >> Yeah. So, we're talking about the courts. Uh, and we've talked about conferences like Comic-Con saying they're not going to allow them anymore. Uh, we had an email uh up from someone saying for the first time they were asked by a security guard at a sensitive area, do you have any smart glasses on? Uh, in other words, they they weren't allowed to bring in recording devices, so technically they were already banned. But like actually acknowledging that you might have them on and we're going to try to make sure you don't have them on has become a thing. Uh so in response to all of this, Brian H wrote, "I am super interested in autofocusing glasses, EXI eyewear, for example. Not having to get new lenses every year would be amazing. Policing these will be harder as the new tech comes out and makes you wonder how long it will take for regular glasses to go the way of the corded phone." Uh, this is interesting because there's different takes on this, but they all use a sensor. >> They all have to have some kind of camera sensor to see what's coming in and then adjust the focus for you. >> So, how do you prove that? Like, no, this sensor is just adjusting my focus, not taking pictures of all the confidential information in your >> Yeah. random random uh guard at the front entrance of the you know the concert hall is not going to immediately understand care about the difference between those two things or whatever >> or the military research room like the concert hall guy may or may not care that much, right? Depends on how much he's getting paid and what mood he's in. The guy at the at the military installation isn't going to let you in. >> Like that's just not going to happen. >> That's so interesting. But yet as a glasses wearer, I mean I wear contacts most of the time, but as a glasses wearer that would be really useful. >> I and I think these are they have a very good chance of becoming a regular product that all of us would love. Like can you imagine if you can go to the eye doctor and the eye doctor's like, "Oh, let me just upload a new prescription to your glasses." And then >> the autofocus or or some of these are good enough that they could inform the eye doctor like, "Yeah, it seems like your prescription is changing a little bit." And then your eye doctor visits become more about eye health and less about assessing what your actual prescription should be. Uh I I think there's a lot of positive things that could come from these. >> Fascinating. >> And obviously, as we talked about in the past, this is going to become a thing like phones where for a while people are like, "Well, you can't have a camera on a phone." Well, I'm not going to get a phone with a camera on it because I won't be able to do this or do that. Or people saying like, "You can't bring your camera into the theater because you might use it to record." uh we're going to get to a place where we normalize these kinds of glasses and there will be policies around them and it will be, you know, they they'll watch to make sure that you're not using them to record, but it's going to take it's going to take a few years to get there. >> I think so. Yeah, I do. >> Uh well, thank you for this. And if uh any of you have thoughts on any kind of story we have, especially if you got some inside information, not like insider secrets, but I mean like your expertise is like, well, I understand why that is. Uh please share it with us. feedback at daily techchnewshow.com. >> Yes indeed. Thanks to Brian H for contributing to today's show. Thanks to you for being along for Daily Tech News Show. You can keep this show in business by becoming a patron and it's real easy. patreon.com/dtns. We'll see you tomorrow. [music] >> The DTNS family of podcasts helping each other understand. Diamond Club hopes you have enjoyed this broker. [laughter]