TNW 434: Is This Framework's Linux MacBook? - Framework's Take on the Pro Laptop

Coming up on Tech News Weekly, Amanda Silberling subbing in for Emily Forlini this week. We talk about ChatGPT's new image generation model. We also talk about how big tech is pushing back on stricter emissions rules. I talk a little bit about the Mythos AI model from Anthropic and Unauthorized Access before Sean Hollister rounds out the show with his initial thoughts on Framework's new Laptop 13 Pro. podcasts you love from people you trust this is this is tech news weekly episode 434 with amanda silberling and me micah sergeant recorded thursday april 23rd 2026 is this framework's linux macbook hello and welcome to tech news weekly the show where every week we talk to and about the people making and breaking the tech news. I am your host, Micah Sargent, and I am joined this week. Normally, we would have Emily Forlini with us, but as I mentioned before, Emily is on maternity leave, and therefore, we are joined today by the wonderful Amanda Silberling. How are you doing, Amanda? Hello, it is me. I can pretend to be Emily if that is easier on the listeners, but that might, I don't know, I can only be me. And you are a great you. This, of course, is the show. Yay, this is the show where we kick off with talking about our stories of the week. These are the stories that we find interesting, that we think you'll find interesting, and that we want to share with all of you. Let's kick things off with your story of the week, Amanda. Yes. So earlier this week, ChatGPT released a new images 2.0 model, which as its name suggests, it's images, but 2.0. And to my surprise, this is actually a really big leap, which I think matters to people because we need to know what level of accuracy AI models are producing images at because it is escalating really, really quickly. And one of the things I wrote about was like two years ago, I wrote an article that was basically about like why AI is really bad at spelling. And in that, I was trying to make examples of Mexican food menus and various different AI image generators and talked to some AI researchers about how the reason why AI spelling is so difficult is because the diffusion models, which most of these models were, just aren't really built to understand. like how do you reconstruct the tiny little pixels that make up text and it just it it like didn't really work um and then chat gpt slash open ai will not tell us exactly what kind of model is behind uh images 2.0 but i would guess that it is something different because now it can spell and it can spell really well and it can spell in multiple languages and i'm worried about society. So yeah, let's, let's kind of, cause I remember too, some of the early days of sort of trying to get, in fact, I remember on this very show having some of the original engineers of Dali on and talking to them at the time about what you were able to create. And back then you'd type something in and the strangest creations would appear. They were very much sort of otherworldly. They didn't make a lot of sense, but it gave this sort of strange alternate reality kind of fantasy. It's like what I imagine the creatures in the wheel of time to sort of have in their universe. And so I remember typing in something like gummy bears having a religious ritual at Stonehenge. Of course. Just to see what would happen. Yeah, you know. And the coolest looking Muse album cover basically appeared in front of me. And I thought it was amazing, but it was nothing of what I had said, right? It's not what I actually wanted. And I think that's what we're just now getting. And that is weird because yeah, as you've said up to this point, at least for me, when I would do it, it was sort of like just a fun toy, right? To play with. Now it's becoming more of that tool that we've talked about before. I popped in a simple prompt to OpenAI and its new image generation. Mind you, I had no text, or excuse me, no text. I had no photos or anything like that. I just said, create the poster for an irreverent pop culture podcast. I said, I want you to name it. Featuring hosts Amanda Silberling of TechCrunch and Micah Sargent of Twit. you are entirely in charge of the theme, the graphics, the text, etc. And it did come up with a show for us called Prestige Trash. Pretty good name. Which is great, which is great. Yeah, irreverent pop culture podcast. And I think the big thing that stuck out to me was the fact that, well, other than it thinking that you are Emily, which is really interesting given that Emily was to be on the show today. I'm like, what is it now? Yeah, right? But as far as I can tell, all the text in here is properly spelled. And it's got references to things. It's doing that cute, what is that called? Half tone. I can't think of exactly what that's called. Yeah, the kind of comic book-y. The comic book-y deal. Yeah. And honestly, although neither of us are really us, I think I look more like myself, but I still don't quite look like myself. And you really don't look like yourself. I'm looking at the image you texted me because we don't have it up on the thing, so we can describe it. But there we go. There we go. Yeah. I mean, I don't look nothing like me. I know that's the scary thing. It does have touch. It's a little bit you, but. It's like this is just another white woman with brown hair. But the thing is that there's a lot of white women with brown hair. There are quite a few brown or white women with brown hair. Yes. But OK, that's like you kind of look like you like. Yeah, I do. Like that does like if I saw that out of context, I'd be like, that looks like Micah. Especially from a distance. It definitely looks like me. And I totally would do that funny finger guns thing. That's so stupid. I would absolutely do that. But OK, the point is here, though, that I typed in that text again with no other context other than the context that ChatGPT might have already and no photos. And it was able. Yeah, it really likes to show my forehead wrinkles. That is for sure. But anyway, the point is that it is it was able to do this with nothing other than a little bit of text and spell everything correctly. And that, at the very least, is new. Now, what I have heard is OpenAI, I don't know that it's been proven, right? But that it is suggested that the company may have used quite a few magazines as a means of kind of taking in more image data. And in particular, where more do you see slick graphics combined with copy, with text, than in magazines? That is, that's a good training place, I think. Yeah, I guess I'd be surprised why that hasn't happened earlier. Because, like, you can get a lot of magazines on the internet. And we know that OpenAI sure does love to scrape the internet for content. That's a good point, yeah. that of course is just one theory about kind of what's going on here but it is it is interesting to see the way that it is kind of moving through things I remember before trying to I had to generate something that I actually was going to use I don't remember what it was regardless I generated it and it created what I was after but there were things that were off about it. And so I was specifically focused on, you know, oh, can you fix the necklace that you put on this person? It doesn't need to have a necklace. And then it would go forth and do everything else except for remove the necklace. And so it's, I don't know, do you feel like AI is still a black box system where you just, where there, it's not clear how to, like what if I pull this lever it does blank or do you feel like we have gotten to a place where for the most part it does what you expect it to do um I don't know I think there's always a degree of variability and I guess like we can't necessarily fully feel confident in how it's going to work when we don't know how it works. And especially with images 2.0, like I was on a press briefing with a bunch of other journalists and one of the other journalists was like, is this an auto regressive model? Is this a diffusion model? And they just wouldn't say. Really? Open AI, very open. That's interesting. Yeah. Which like, I mean, I am not an AI scientist, but like the vague general overview I have, I'm like, it would make sense if it was something that's not a Diffusion model. But like, I don't know. I mean, when I had it make a Mexican restaurant menu, I found like one small error. Like, but if I got that menu, like at a restaurant today, I would not necessarily feel like something was off, except maybe that like carne asada isn't spicy and it was listed as spicy on this menu. But I don't know, maybe this chef really likes to put more of a kick in the carne asada. It could possibly be that. We see companies working toward releasing new features, AI companies specifically, working toward releasing new features all the time. And it has been interesting seeing how those features and the sort of announced features and the competition has evolved, where at one point, one might argue that Anthropic was behind because, for people who might not know, Anthropic is the maker of Claude, and may have thought at one point that the company was behind because it didn't do image generation. and now that's really like we've we've almost gotten to a place where you don't necessarily need to do all the things and in fact open ai has reshuffled its priorities refocused on certain things and stopped doing sora the video um social media deal r.i.p and has kind of focused on the code aspect of it. But Anthropic's claw just completely says, no, we're not doing visuals other than those generated through vector. There's a new design tool from Anthropic. But my point there is, at one point, all we were doing was seeing a company announces a thing, then the other AI company announces that thing. And then the third AI company goes, oh yeah, it took us a while, but we're there too. And it's becoming less, I think, about feature parity now. And finally, about diversification. How is my AI better than yours? Not how is my AI the same as yours? So you know, you're getting the same thing across the board. And that I think is cool because that is, in my mind, a better form of competition within tech than simply having feature parody. So, yeah. It is time for us to take a little break before we come back with my story of the week. I want to tell you all about WebRoot bringing you this episode of Tech News Weekly. If your computer feels sluggish, if it heats up when you open a few tabs or if it sounds like it's preparing for liftoff every time it runs, your device probably isn't the problem. Your antivirus may very well be because many big name brands have become bulky, complicated, full of pop ups, full of upsells. WebRoot offers an all in one digital protection for up to 10 devices with a variety of plans designed to protect you and your loved ones from digital threats. You get powerful antivirus and identity protection without the slowdowns, without the pop ups. WebRoot keeps you protected online while staying out of your way. WebRoot Essentials scans six times faster and takes up, I love this, 33 times less space than the average competitor. And it ranks number one in performance compared to Norton, heard of them? McAfee? Yeah, you know. WebRoot Essentials versus Norton Antivirus, it scans 3.7 times faster, installs 35 times smaller, and is five times less RAM when idle. WebRoot Essentials vs. McAfee Antivirus, it scans 10 times faster, installs 16 times smaller, and is also 5 times less RAM when idle. These differences can make your computer feel newer, faster, and easier to use. WebRoot also offers WebRoot Total Protection, which includes antivirus, as well as identity monitoring, privacy protection, and cloud backup, all in one simple, hassle-free subscription designed for everyday life. WebRoot Total Protection ranked first overall when compared to top competitors, scans seven times faster than the average competitor, and takes up three times less space than the average competitor on the hard drive. AI has completely changed the cybersecurity game. Scams are smarter, malware is faster, phishing emails look shockingly real, but the good news is you don't need to be a tech expert to stay ahead of it when you use security that can keep up with AI threats. Unlike free antivirus tools or older security programs, WebRoot is built to counter modern AI-driven attacks. It's fast, it's lightweight, it's designed to spot threats before they even reach you. So live a better digital life with WebRoot. WebRoot is offering our listeners an exclusive 60% off offer. Visit webroot.com slash twit to learn more. That's webroot.com slash twit. And we thank WebRoot for sponsoring this week's episode of Tech News Weekly. All right we are back from the break joined as I mentioned by Amanda Silberling this week And it is time for my story of the week because some of the biggest names in tech are pushing back against proposed changes that would make it harder for them to claim they running on clean energy In a new report from Bloomberg's Olivia Ramond, we learn that Apple, Amazon, Meta, General Motors, FedEx, BYD, and Patagonia are among more than 80 companies that signed a joint letter to the Greenhouse Gas Protocol. Now, the Greenhouse Gas Protocol is the global standard center for corporate emissions reporting. And in this letter, the companies objected to a potential tightening of the rules that govern how they report their scope to emissions. At the heart of this debate is whether companies should be allowed to kind of keep buying these renewable energy certificates from distant wind and solar farms to offset the electricity that they actually pull from the local grid. or whether they should have to match their clean energy purchases more closely to when and where they're actually using power. It's kind of a technical sounding fight, but it has real consequences on how credible corporate climate claims actually are. Now, let's talk about this because essentially right now you can do a bunch of energy, use a bunch of energy, and then a long time can pass and then you can buy credits to offset that energy. and what this is doing what we see happening in some cases and i think why there's been this push for tightening potentially is because the companies will do their thing and then at the end of the year as the year starts wrapping up they'll buy credits to offset but as these folks are pointing out the problem with that is when you're buying your credits all at once or you know within a period of time that is far removed from the time that you actually done the thing, A, it isn't the same thing. It's more just about throwing money at the problem. And B, oftentimes it means that you are getting your credits essentially from places that are outside of the grid that you're actually having an impact on. So the argument there or the wish there is for companies to have to get, if they're going to do credits, that they have to get their credits sooner to the time that they're using the energy. And then they also have to get their credits locally or as close to local as possible so that it's not just about throwing some money at the problem far away and not having an impact. The Greenhouse Gas Protocol, it's a global framework that's overseen by two nonprofits. First, the World Resources Institute and the World Business Council for Sustainable Development. These scope two guidelines haven't been updated in more than a decade. So that does kind of make sense that it's worth doing. And the current revision process closed for stakeholder comment at the end of January. The proposals on the table would change everything. Most companies currently buy renewable energy certificates called REX, again, on an annual basis. But the math is loose. A company can purchase certificates from a wind farm in one part of the country to offset the electricity it pulls from a fossil fuel-heavy grid somewhere else at a completely different time of day. So companies would need to instead do hourly matching and geographic proximity so that the clean energy, it's coming from the same place as it's being used. It's because many companies can't actually run on renewable electricity around the clock. Solar power, of course, drops off at night. Wind, it's not always windy. And most corporate operations are pulling from a mixed grid. So the annual certificate system lets companies claim 100% renewable status, even when their real-time electricity uses anything but. I wanted to talk to you because, Amanda, I'm sure you have seen, as we've covered tech companies, a lot of companies touting their environmentally friendly policies. In the past, I would say that since the change in administration, it is not as front of mind. We haven't seen Lisa Jackson talking to Mother Earth in an Apple announcement video in a while. And so these things are kind of, if not backseated, they're sort of subtle. And that said, though, I feel like companies wear this as a badge. Does it impact the way that you view the companies when you see the green policies sort of being put forward and shared and talked about? Does it make a difference for you? i mean i'm cynical so i feel like when i see like apple talking about how environmentally friendly it is i sort of take it with a grain of salt but i think that is in part because of some of these issues like the fact that like buying carbon credits it's like it doesn't do nothing but there's a lot of doubt about just how effective that is and it's kind of like a get out of jail free card almost except um get out of trouble for uh using uh energy in an inefficient way by paying money which you have an excess of free card but um wow you know maybe we'll have to have chat generate that card for us yeah but like i don't know i mean these sound like reasonable things to me where like yeah like if you want to buy carbon offsets if you want to like say that you are operating on renewable energy then just sort of throwing money at it which is how i feel carbon credits have been working in the past it doesn't really make sense and it's really easy to kind of fudge the numbers or like I mean, they are literally just like throwing money at the problem. Yeah, I mean, that's exactly what it is. And I think that's exactly what the commissions are trying to avoid. Now, the letter that was signed by Apple, by Amazon, by Meta, by GM, on Dash or on Dancer, on Prancer, on Vixen, argues that the proposed changes would make these guidelines optional rather than mandatory and warns that the revisions could slow corporate investment in clean energy projects. Apple said that it's not, or well, it didn't say it outright, but raised concerns about feasibility saying, quote, we have concerns about the, well, literally in the next sentence, we have concerns about the feasibility of implementing the proposed hourly and location matching requirements, not only for our own scope two, but for our suppliers who contribute to Patagonia's scope three footprint. This is senior, excuse me, senior Environmental Impact Program Manager for Decarbonization at Patagonia. Schenck added that he believes the proposal, as written, will have a chilling effect on investment in renewable energy. But of course, you know, others are not quite buying into the industry's concerns. Matthew Brander, who's a researcher at the University of Edinburgh, who sits on the GHG Protocols working group said this, some people want the rules to be more robust and have more integrity and other people just want the status quo. So, you know, fair. This is, I think that it's a fair suggestion, right? That simply waiting until the end of the year and paying some money to get to claim that you are being renewable, that you're being this, that you're being that. It's not, I don't know that it should have existed in the first place as an option, but the one thing I'll say is perhaps it helps to build infrastructure in those places where those credits were going. And so that's a good thing. It's having a positive impact there, but it's just not ultimately what we would want from these companies. And it's one of the things that you hear from people who study sustainability, who study all aspects of sustainability, including recycling and everything, is that so much of the time we individuals take on the guilt and therefore the responsibility, right, of, you know, you separate out your number two plastics from your number seven plastics and we recycle and we do this and we do that. and it is but a drop of a drop of a drop in the bucket. It is corporations that need to be making these changes. And that, I think, is once again used as a means of getting rid of that responsibility and saying, this isn't for us to worry about. You should be recycling. Don't forget to put it in the blue bucket. And it's like, yes, we should do that. But for meaningful change to happen, the corporations need to be making those changes. And I feel that GHG and the other group are on to something with this personally. Yeah, I mean, I guess I sort of think about it like I think that in the Silicon Valley universe or like people that would agree with the like tech accelerationist movement, there's this idea of just throwing money at tech and seeing what will happen. and that like we should just like let these companies cook because they're going to figure out climate change and whatever but like that's so dubious and I feel similarly about like there are some problems that believe it or not just throwing money at it doesn't solve like if I break my leg and then you give me ten thousand dollars I'm like cool thanks for the $10,000, but my leg is still broken and I have to wait for it to heal. And like, it's like, yeah, like the $10,000 helps, but like it doesn't fix my leg. Yeah. That's, that's a really good point. Well, I want to thank you for taking the time to join us today on the show. It's always a pleasure to get to chat with you. If people would like to keep up to date with the great work that you're doing, where should they go to do so? I'm mostly on blue sky at amanda.omg.lol, a good URL that I own. I haven't been on X, but I have been on X this week solely because people are saying that I'm funny there. So, um, yeah. So everyone go and tell Amanda that she's funny and then you'll be in the places that they are. That's good. Yeah. Yeah. I, yeah. I wrote an article that people think is funny. And then I was like, yay, this is fun. And then I started seeing more like random sexist stuff on the timeline and then it was less fun. And you're like, yeah, this is no longer fun. Yeah. Otherwise, find me on TechCrunch. Find me on my podcast, Wow If True. And here sometimes more often than you would think. Oh, and the upcoming Prestige Trash. Yes. Our new podcast, Prestige Trash. Thank you, Amanda. We appreciate it. Yep. Bye. Bye-bye. All righty, folks, let's take a quick break before we come back with another story and then an interview after that. I want to tell you about OutSystems, bringing you this episode of Tech News Weekly. OutSystems, the number one AI development platform. See, OutSystems will help businesses bridge the enterprise gap to their agentic future, where the constraints of the past give way to unlimited capacity and scale. Outsystems enables them to build AI agents that can actually do work. So take actions, yes. Make decisions, yes. Integrate with data rather than just answer questions. Outsystems provides the only AI development platform that is unified, that is agile, and that is enterprise proven. When it comes to the unified aspect, you can build, you can run, you can govern apps and agents in one platform. As far as agility, you can innovate at the speed of AI without compromising quality or control. And enterprise proven, yes, it is trusted by enterprises for mission-critical AI applications and durable innovation. Outsystems, it's the secret weapon behind the world's most successful companies. They're not just for small apps. They're for the massive, complex systems that run banks, that run insurance companies, government services. Outsystems even helps companies with aging IT environments bridge the gap to the AI future without our rip and replace nightmare. Outsystems provides the safest and fastest way for an enterprise to go from we need an AI strategy to we have a functioning AI application. Yes. Stop wondering how AI will change your business and start building the agents that will lead it. Visit Outsystems.com slash twit to see how the world's most innovative enterprises use Outsystems to build, deploy, and manage AI apps and agents quickly and cost effectively. without compromising reliability and security. That's O-U-T-S-Y-S-T-E-M-S dot com slash T-W-I-T to book a demo. And of course, we thank OutSystems for sponsoring this week's episode of Tech News Weekly. All right, there was a little break from the AI, but it is time to run back into the fold here because I want to tell you about Mythos, a little more about Mythos. As we saw a report from Bloomberg, Anthropic has this new AI model. Anthropic has this new AI model that's so capable of finding and exploiting software vulnerabilities that the company says it's too dangerous to release to the general public. Fortunately, unfortunately already, a small group of unauthorized users has managed to get their hands on it. That's the reporting from Rachel Metz at Bloomberg, who broke the story on April 21st. And that's two days ago as we record this on the 23rd. The model is called Claude Mythos Preview. And according to Anthropic, it can identify exploitable flaws in every major operating system and web browser. In fact, I encourage you to go watch the recent, I think the last two episodes of Security Now to learn more about it. But the point is that it identifies these flaws in so much of the software that we have. And Anthropic said, look, we're only going to make this available to a hand-picked group of software providers with our special initiative that they call Project Glasswing. But on the same day that Anthropic announced that limited release, a group of users in a private online forum had already found a way in and have reportedly been using Mythos regularly ever since. Rachel Metz reports that a small group of users in a private Discord gained access to Mythos on the very same day that it said, hey, these are the companies who are going to be testing it. The person who's familiar with the matter, who of course asked not to be named, told Bloomberg that the group has been using Mythos regularly and corroborated the account with screenshots and a live demonstration of the model. Now, this does matter because Mythos, it's not just another sort of latest and greatest model right Anthropic has said it not going to be releasing Mythos broadly because of the fact that it can be used in cyber attacks Now what I understand is that the group did not choose to use this for cyber attacks and anything in that field because they didn want to alert Anthropic to the fact that they had access the tool But this, of course, the whole premise of Project Glasswing is that this version, Mythos, gets shared with vetted partners, Apple, with Amazon, with Google, with Microsoft, with CrowdStrike, with the Linux Foundation, and that these companies are able to then run it and see how there may be issues with their tech. but it's tightly controlled and supposed to be used for defense, not offense. The unauthorized access, of course, punches a hole in that premise and it did so on day dot. Now, how in the world did the group get in? How did, how did Mythos not pick up on the fact that this was going on? Well, I think that this is arguably the more interesting aspect of the story. According to Bloomberg's reporting, there were sort of several ways that the company did it. One member of the group was actually working as a contractor at a third-party vendor that Anthropic uses. So it's sort of two times removed, right? The group used commonly used internet sleuthing tools. So those that, you know, cybersecurity researchers use regularly. And they're part of a private Discord channel that's focused on hunting for information about unreleased AI models. They used bots, to scour GitHub to scour other unsecured websites for details Anthropic and its partners had posted. Now, to actually locate Mythos, the group made an educated guess about the model's online location based on knowledge of the format that Anthropic has used for other models. So, yeah, I suppose if you have enough information about how Anthropic is storing and where Anthropic is storing, you suddenly have the opportunity for pattern recognition. And when you have the opportunity for pattern recognition, you find a pattern. And that pattern often will lead you to the truth. You have solved my dungeon puzzle. As a source told Bloomberg, such details were revealed in a recent data breach from Mercore, an AI training startup that works with a number of top developers. Here's what Anthropic had to say in response to the Bloomberg report. said, we're investigating a report claiming unauthorized access to Claude Mythos preview through one of our third-party vendor environments. The company also said it currently has no evidence that the access reported by Bloomberg went beyond a third-party vendor's environment or that it's impacting any of Anthropik's system. So what's the group actually doing with this? How are they using Mythos? Well, apparently, the person familiar with the matter told Bloomberg that the group is interested in playing around with new models, but not wreaking havoc with them. They haven't been running cybersecurity-related prompts on Mythos. Instead, they've just been trying to use it for simple tasks, build some websites, other simple things. I don't know. Show me the ingredients in a hot dog bun. I don't know. But part of that is just to avoid detection by Anthropic. The source said that the group has access to a slew of other unreleased Anthropic AI models and said that the person has permission to access Anthropic models and software related to evaluating the technology for the startup they work for. So they actually gained this access from a company for which they had performed contract work evaluating Anthropic's AI models. Bloomberg chose not to name that company for security reasons. It gets complex when you have so many different contractors working for so many companies that are contracted and then need more people. So then they contract out. It gets really messy. And I think this is why it's important to understand what Mythos is, what it can do. It already found thousands of zero-day vulnerabilities, a 27-year-old flaw in OpenBSD. It's kind of wild. And so even if this group chooses not to use the tool in an irresponsible way, the fact that it can is a bit of an issue. The Mythos situation highlights kind of some tension that exists at the heart of Anthropics approach because the company, of course, like other AI companies, wants to continue to develop increasingly powerful AI, but it aims to keep that technology from spreading beyond its approved partners. And Anthropic in particular seems to have and hold to this idea of creating AI for good and responsibility and safety. That's sort of part of what the announcement of Mythos was. It was a bit of a signal being sent out that we can make very powerful AI, but watch how we responsibly do it. So the fact that this has happened certainly makes the claim feel a little less true. Of course, Anthropic continues to keep an eye on the situation. And luckily, journalists at various online publications do as well. So we'll continue to monitor this for future outcomes there. In any case, if you can believe it, it is time for us to take another quick break before we come back for our interview on today's episode of Tech News Weekly, talking about a very cool laptop that is unique in the space. But first, let me tell you about Zscaler, bringing you this episode of Tech News Weekly. Zscaler is, of course, the world's largest cloud security platform. The potential rewards of AI. You know it at this point. The AI companies know it at this point. The rewards of AI are too great to ignore. But like they say, with great power comes great responsibility. The risks, the loss of sensitive data, the attacks against enterprise managed AI, that is also two greats to ignore as well. Generative AI increases opportunities for threat actors. It helps them to rapidly create phishing lures that are customized. You get malicious code written. You get automated data extraction. In fact, there were 1.3 million instances of social security numbers leaked to AI applications. ChatGPT and Microsoft Copilot saw nearly 3.2 million data violations. So it's time to rethink your organization's safe use of public and private AI. Check out what Siva, the Director of Security and Infrastructure at Zura, says about using Zscaler to prevent AI attacks. With Zscaler being in line in a security protection strategy, it helps us monitor all the traffic. So even if a bad actor were to use AI, because we have a tight security framework around our endpoint, helps us proactively prevent that activity from happening. AI is tremendous in terms of its opportunities, but it also brings in challenges. We're confident that Zscaler is going to help us ensure that we're not slowed down by security challenges, but continue to take advantage of all the advancements. With Zscaler Zero Trust plus AI, you can safely adopt Gen AI and Private AI to boost productivity across the business. Their zero-trust architecture plus AI helps you reduce the risks of AI-related data loss and protects against AI attacks to guarantee greater productivity and compliance. Learn more at zscaler.com slash security. That's zscaler.com slash security. All right, we are back from the break, and that means it's time for the interview. Framework has been building a reputation for modular, repairable laptops. But its latest release, the Laptop 13 Pro, arguably takes things up a notch. Is this the machine that finally lets Framework go toe-to-toe with Apple on build quality? Joining us to break it all down is The Verge's Sean Hollister. Welcome back, Sean. Thanks for having me back. Yeah, pleasure to have you here and also very excited to hear about your experience with this device. First and foremost, could you tell us a little bit about kind of the key upgrades that make the Framework Laptop 13, long name, Framework Laptop 13 Pro a significant leap from the company's previous models? And I think the bigger question for me is, what is it that makes Framework sort of position this as the MacBook Pro for Linux users? Yeah, so it starts off with the frame itself. Framework has always been about building laptops that are very modular, very upgradable, but they haven't kind of nailed some of the basics that you think about when you think about a premium laptop. Most of the premium laptops out there, they are made of a block of aluminum that has been CNC machined by a computer-controlled machine into a particular shape that feels very rigid when you pick it up. It's fairly lightweight because it's aluminum, but very rigid, firm. It has that monolithic appearance and feel to it. And theirs had not been that. They'd been stamped metal with plastic bits in there to kind of hold it together. They looked cool, but when you feel them, they don't feel quite as premium as your MacBook. This is that thing now. It's using 6000 series aluminum, machined out of the blocks. When you pick it up, like I picked it up and tried to twist it, and it felt solid. You push down on the keyboard, and you don't feel give under your fingers. You feel like you're just typing on nice keys. The trackpad, they've spent a lot of effort into the trackpad. I don't know yet if it's MacBook beating. It's certainly not as large as some of the bigger MacBook trackpads out there. But this trackpad feels buttery and smooth. And it is this haptic type of trackpad that you get used to with machines now where there aren't physical buttons. I know some people still like the physical buttons. But you can just touch the trackpad and you still feel a response as if you're pressing down on a button. Wow. That's just the start. Yeah, I was going to say that already is really cool. I mean, you did, as you're talking about here, get this hands-on time with the 13 Pro. Is there anything else that stuck out to you when it came to trying this laptop? I think in particular, kind of maybe what were your expectations going into it? And were those at all changed by what you saw? particularly when we're talking, at least for me, when I think about something that I can make changes to that is modular, that is adjustable, I'm immediately going to give an item grace in terms of its fit and finish. But it sounds like framework has figured out both. So yeah, could you tell us about that expectation and then the outcome that you saw there? As a journalist, I try not to be fans of companies. I've always, always in all my life, I don't buy company. I don't buy products from companies because I like the company. I look for a particular product that has been done very well. And I buy that product after I read the review. And so I don't want to like jump ahead and be like, you should buy this laptop necessarily. but both on the spec sheet side of things wow this has a lot of things i'm looking for and the actual feel when you get there and you touch and play with it quite amazing and i am kind of a fan of framework because they are the only company that has ever nailed this idea of you can buy a machine and you can upgrade it in a year or two or three with a new chip motherboard memory everything that fits inside the chassis you can take a new motherboard and stick it into an old one and it will actually give you that upgrade without having to buy a whole new laptop companies have really never done this before there have been a few that have promised to do this and they've abandoned this promise after a year you know you could upgrade it once the next year and then never again there are some where you could many where you could have upgraded the memory but You don't get that from Apple machines anymore. You can't upgrade the memory or storage in Apple machines. You can't in the framework. They'd already nailed that kind of thing. And so when I got the spec sheet for the 13 Pro here, I was excited about, oh, it's going to have a fully custom, you know, 2.8K variable refresh rate screen, 30 to 120 hertz. Finally, a smooth screen on a laptop. That'll be great. I was excited to hear it'll have longer battery life, maybe 24 hours of 4K Netflix from this new Core Ultra Series 3 chip, I was excited to hear it'll have this compression-mounted LPCAM2 memory, where you put the memory, you screw it down to the motherboard, but it is still replaceable. But because it is tightly paired to the motherboard, it has more direct link than kind of the socketed memory you've had in the past. It's more fast and more power efficient. All these things, all these specs in my head, I'm like, yes, this is going to be a great laptop on this spec sheet what i didn't expect to be webbed by is that build quality i told you about them before i'm like lots lots of companies lots of companies have have done machined aluminum chassis you know you can get that from asus you can get that from asus you can get that from a lot of companies but because framework had had the one side of things but not the other but not the creature comforts i was skeptical that they'd put all that together and nail it. So now I'm like, wow, the build quality is there. I can see that right away. Now I just wonder if the spec sheet is going to deliver in the full review. Is this going to have the performance of a MacBook? I don't know. Yeah. I mean, so you, you mentioned that display, right? The 2.8K display, color gamut capped and perhaps some, some moments of, of trade-offs. Could you tell us what Framework told you about those tradeoffs when it came to the screen in particular? Yes, yes, yes, yes. So Framework here, this is the first fully custom display that it's put into a laptop. Before it has had, you know, we've taken an off-the-shelf display. We've modified it in a couple ways. But generally, they've been careful to find a part that is available for a good price, that has some good characteristics, and that they think they can have be upgradable down the road. You know, they need something that's going to fit into the form factor down the road. If they make a laptop with a, you know, a 16 by 9 display at a certain, you know, inch size, they need to make sure that years from now they can get that size and aspect ratio of display. this one is a three by two display so it's a bit taller which is nice it's brightish at 700 nits not the brightest out there and like you were mentioning there is something about the color so they're doing a hundred percent of the sRGB color space and they are color calibrating each screen on each one of these machines so that's super nice that they're color calibrating but you think a hundred percent color calibrated that's going to be great for creative professionals right who need this for the images well 100 sounds great but of the s rgb color space that is one of the smaller color spaces out there when it comes to the kind of colors you need to be able to see on a screen to do very creative work uh you want generally want to hear 100 of adobe rgb you might want to hear 100 of dci p3 as your color gamut let's be like oh yeah that has all the colors i need to be able to see on it So this it sounds like a really nice screen It looks like a really nice screen in person for gaming I be fine with it for productivity I be I be great with it I just don't know if, if you're going up against the MacBook Pro for creative professionals, I don't know if that's the screen you'd be going for. Again though, it's also a 13.5 inch laptop. This is the kind you stick in your backpack and you take places. It's not necessarily the one that you're going to have on your desk all day, every day. So you could still pair this with a different monitor that has those characteristics you're looking for. They said what they're going for here is the ultimate developer laptop, not the ultimate laptop for creative professionals. Pro means developers here. That makes sense. Now, I think another aspect of this that I was kind of blown away by, frankly, is that, correct me if I'm wrong, existing laptop 13 owners can update some of those components. Tell me more about backward compatibility for the company here, especially when you are having to make these trade-offs, right? Between let's do a display, but we need to customize it slightly versus a fully custom display. But our whole thing is that we allow people to upgrade. That's hard to do. And so hearing that there's like any level of backward compatibility is fascinating to me. Yeah, there's there is quite a bit. I mean, it is not 100% backwards and forwards compatible, but it is almost there. How it works is with with every one of these framework laptop 13s, you can start with the ports. Their thing from the very beginning is you can pick your own ports, you can take the four little expansion cards on the side of the machine with a HDMI port, with a USB port, with a USB-A, with USB-C, headphone jack. And at any time, pull one of those out of the side machine and replace it with a different port for the task that you want, SD card. But they do this kind of thing inside the machine too. You want a new display. You can drop in a new display. You want a new Wi-Fi card. You can drop that in. You can drop in an entire new motherboard that includes the memory and the new processor on it. You can drop in the new keyboard cover and touchpad if you want this haptic touchpad. The one thing they had to do is because this is a different kind of frame, the actual chassis is different than the previous laptop 13s, you need to buy a few of those components as kits. So if you want to, say, get the bigger battery, you'll also need to get the bottom chassis kit, I believe, to do that. And that also comes with, they call it a bottom cover upgrade kit. So you replace the bottom cover of your laptop up with this machined aluminum one and that has space for the bigger battery and your uh your new motherboard with processor and everything in there uh there is also a um uh the new with the new touchpad and keyboard you need to buy that as a set it is one kind of like top lid that goes uh on one lid that goes on top of the base of your computer and that has the keyboard and the touch pad in it so you buy that whole piece but if you buy that whole piece you can add that to an existing laptop and screw it in so you're like if there's one thing you didn't like about your 13 you can probably use the 13 pro version of the part and fix it or multiple ones of those things you want to get a a nice translucent bezel for your screen just customize it make it look cool that's super easy, just snaps right on there. Anybody can do that. Other things, a few screws, you know, anywhere from five to, you know, 20 screws, depending on whether you just want to replace the cover or whether you want to like pull out the entire motherboard and upgrade all the internals. And we, you talked a little bit about this memory, right? This seems to be playing a big role and framework is at least being more candid and upfront than we've seen from some other tech companies about the RAM crisis. How is the company handling this? And what did the company have to say about sort of how things are with RAM right now? Yeah. Framework always does its products in batches so if you want one of these new laptops you sign up for you pre-order and you get put into one of its batches and it'll be like oh your batch is going to ship stop sorry about that your batch is going to ship in june oh your batch is going to ship in july if you're in batch you know five or something like that it keeps moving out as you go so they're not saying we have all the ram in the world to fill everybody's laptops right away but they are saying they are working directly with the memory suppliers like Micron, you know, CEO Neera Patel, name drop Micron to me to make sure that they have this kind of memory. They've been working together with Intel, kind of the CPU and the memory here is a bit of a package deal. They need to link to each other very directly through this motherboard. And so they have support of bigger names to make sure they get memory for these machines. I don't know what they do if everybody decides that they want one, but so far not everybody says they want to spend extra money for a laptop that you can upgrade this way. All right, I got to pop in here for just a quick moment so I can tell you about Melissa bringing you this episode of Tech News Weekly. Melissa is, of course, the trusted data quality expert since 1985. 84 percent, that is the percentage of organizations who struggle with inaccurate or duplicate data, which impacts everything from fraud prevention to AI performance. Melissa has been solving this for going on 41 years now. So who better to help you out than Melissa? Here is what that means for your business. You get global address verification, which validates and standardizes addresses across more than 240 countries. You get mobile identity verification, which matches customers to mobile numbers, cuts fraud, opens SMS channels. On average, a database contains 8 to 10 percent duplicate records. Melissa's matchup technology can identify non-exact matching duplicate records that That would traditionally be missed. You can enrich records with demographic, geographic, psychographic, firmographic, and property info for more intelligent targeting. And Melissa Alert Service will monitor and automatically update your customer data for moves, for address changes, for property transactions, for hazard risks, and more. Join more than 10,000 businesses worldwide using Melissa. eToro, the social investing platform with 23 million users, uses Melissa for identity verification. Their business analyst says, We find electronic verification is the way to go because it makes the user's life easier. Users register faster and can start using our platform right away. No matter the size of your business, Melissa integrates with you. Melissa has easy-to-use apps for Salesforce, Dynamics, CRM, Shopify, Stripe, Microsoft Office, Google Docs, and more. Melissa's solutions and services are GDPR and CCPA compliant, FedRAMP and ISO 27001 certified, and meet SOC 2 and HIPAA high trust standards for information security management. Get started today with 1,000 records cleaned for free at melissa.com slash twit. That's melissa.com slash twit. And then the price starting at $1499 prebuilt climbs quickly from there. Higher end configurations are in MacBook Pro territory. uh do you feel that the price to value um ratio the like is it is it reasonably priced based on what we understand about the market right now uh paired with the capability of this device paired with the fit and finish of this device etc etc etc the entry level price at your 1499 for pre-built or i believe it's a 1200 for diy which by the way with framework diy means we send you the laptop almost finished and then you stick memory storage and your own you know os on it it's not a lot of diying there that price feels reasonable for the moment that we are in now with gobs of uncertainty and ram prices being what they are and the kind of functionality and upgrades we're talking about here, they feel like a reasonable amount to ask over the existing Laptop 13 or over some of the competition in the Windows and Linux space. I should mention, though, that at $14.99, you are talking about pre-installed Ubuntu Linux, which is kind of cool, but not Windows. If you're thinking that that is a Windows machine at that price, you do have to get your Windows license separately or configure with that, and you can find a copy of windows relatively cheaply many places on the internet so i wouldn't worry too much about that but it is a thing where that is something you may not expect previously the starting price for a pre-built framework laptop was with windows installed and this is not that in terms of like once you get up there in price i'm not sure it depends on the performance if we play with this and we're like wow this is as fast as a macbook pro or faster and it's got you know this just as good battery life and the build quality is there and also by the way you can upgrade this down the road that value proposition will be excellent and if it's not there if you know if it's outpaced by apple or if it's outpaced by windows competition um it's hard to tell before thoroughly testing a laptop whether it will be as fast as even the name of the processor on it suggests for many years now it's been the case that the measure of a processor is not the the megahertz or the gigahertz or you know the memory or the cache it is how much thermal headroom the processor has and how much battery power it has to do its job consistently any processor on the market now will thermally pull itself back if it's getting too hot and so we have to see what this slim chassis with this fan does it could be excellent it certainly didn't seem very hot when i picked up one that was playing cyberpunk 2027 or 2077 at the event and uh on a call with uh with framework ceo nirav patel he told me we've been talking for the last half an hour uh in the zoom chat on the laptop 13 pro and he said it was barely warm so this could be great we don't know that's nice wait for the reviews yes indeed uh and then my last question for you this is being framed as you know the developer laptop rather than visual designers which i think we're kind of used to the creative but the visual creative um when it comes to to these portable but powerful machines. Based on what you saw and heard at the event, who is, I know I always, when I have you on, we're talking about reviews or summations of what you've seen. It's always the question of who is the ideal buyer, but I'm asking it again, who is this for? This is for people, and this is definitely not everybody, but this is for people who have this amount of money to spend want something that is portable with long battery life and comfortable keyboard and touchpad, who want to do that kind of portable computing, and who think that at some point down the road, they might want to keep using the same machine instead of buying a new one. A lot of people have this urge. Not everybody does, but some people will argue vehemently, no by the time i am ready for this laptop to change i will want a whole new laptop i will have worn out various parts i will be ready but many people aren't i have a logitech mouse here this logitech mouse i love it i love the way it feels in my hand uh if the rubber on it were not literally falling apart and if logitech would sell me a replacement for that rubber piece that is falling apart on the side of my hand i would keep using this mouse indefinitely wow i would never want to change away from this mouse because it does what i need i do not need more responsiveness in a mouse i do not need i do not need or want it to feel differently i just want it to stay clean and comfortable under my hand logitech does not do that right now framework does do that with laptops and so uh if this build quality is is what you've been waiting for and if the performance and battery life or what we hope they will be. That may unlock this for people who wouldn't consider a framework previously because of creature comforts like that. Well, Sean, I want to thank you for taking the time to join us today. It is always a pleasure to get to chat with you and see the cool stuff that you've tried out. If people would like to follow you online and keep up to date with all the great work you're doing, where are the places they should go to do so. Yes, I'm at theverge.com. It's a wonderful website. I hope you'll join us. Maybe subscribe if you want to help support my ability to do this job. And I am on Blue Sky as Sean Hollister. Awesome. Thank you so much. We appreciate it. Thanks for having me again. Alrighty, folks. That brings us to the end of this episode of Tech News Weekly, which means it is time to say goodbye to all of you. Thank you so much for tuning in. If you would like to get ad-free episodes of this show and every single one of our shows on the network, can I invite you to join our club? Twit.tv slash club twit is where you go to sign up. When you sign up for the club, you can, again, head to twit.tv slash club twit or use that QR code in the top corner there. And doing so, we'll point you to a place where for $10 a month, $120 a year, you will become a member of the club and gain access to, as I mentioned, all of our shows ad-free, but also you will gain access to some special feeds that are yours. You will get a feed that has our behind-the-scenes moments. You will also get a feed that has our live coverage of tech news events and a third feed that has our special club Twitch shows like MyCraftingCorner, Stacy's Book Club, Camera Photo Time. We've got coffee stuff on occasion. We've got the wonderful AI user group and so much more. Plus, we've got even more planned soon. So now's the time to join the club. If that's not enough for you, fine. It's I understand. Because, ah, I've got another offering. Discord. Yes, you can join our Discord where you can chat with fellow Club Twit members and those of us here at Twit. And we'd love to see you in the Discord as well. Head to twit.tv slash clubtwit to sign up. If you'd like to follow me online, I'm at Micah Sargent on many a social media network where you can head to chihuahua.coffee, that's C-H-I-H-U-A, H-U-A.coffee, where I've got links to the places I'm most active online. Thank you for tuning in. We do appreciate it. And I'll see you again next week for another episode of Tech News Weekly. Bye-bye. This is not the future we were promised. Like, how about that for a tagline for the show? From the BBC, this is The Interface, the show that explores how tech is rewiring your week and your world. This isn't about quarterly earnings or about tech reviews. It's about what technology is actually doing to your work and your politics, your everyday life. And all the bizarre ways people are using the internet. Listen on BBC.com or wherever you get your podcasts.