SED News：PHPの巨大脆弱性、OpenAIに仕掛けられた盗聴器、そして死んだAIプロダクト

Hello, and welcome to SED News. This is the monthly installment of Software Engineering Daily, where myself, Gregor Vand, and Sean Falconer are with me. Hey, everyone. Hey, Gregor. we take a spin through the tech headlines we then do a sort of bigger topic in the middle and then we take a look at our favorites from hacker news have popped up through the month as well both of us have been pretty busy i think since we last were doing fd news but we did actually see each other in person as well in sf which was very very nice always to see you in the real world not just on a video screen. Yeah, we went, I don't know how long it was, well over a year having never met each other, and then now three times in a couple quarters, so that's pretty good. Yeah. So, yeah, I was over in SF for the Super Bass Conference,

although we kind of call it a festival. I don't know, Sean, you popped your head in for a couple hours. Could you attest to it being more of a festival than a conference? Yeah, I guess it kind of had a festival-type feel to it, but you could probably also label it a conference, but definitely a good energy and vibe there. Sweet. Nice. Yeah, I managed to briefly meet Steve Wozniak, which was quite a bit of a pinch me moment. But actually just seeing him speak on stage was actually great as well. He was very funny. What was he talking about? It's funny, we had our CEO and CTO interviewing him, and they kept sort of trying to ask about AI and things, and he kept sort of changing the topics back to really just his personal life and stories from before, bringing up things to do with Steve Jobs, and took a dig at Oracle, which obviously we quite enjoyed that because he was seeing how much he loved Postgres. So yeah, it was really just sort of whatever he felt like talking about, but that was kind of nice in a way. And it was actually his, he was doing this on his wedding anniversary. So, and he has apparently wine that he has his name on or something.

So they were like drinking the Woz wine on stage. And it was kind of a nice way to round out the conference, I guess. Yeah, that sounds awesome. Yeah, otherwise I got to, I'm sure a lot of people listening have done this already, but riding a Waymo on the freeway, like an early access thing, apparently. I kind of forgotten I'd signed up to early access on my last trip. and then I needed to take a car to the Stripe office, which is down in a place called Oyster Point, which is sort of south of San Francisco. And Waymo claimed it would do it for me. And I was like, that sounds great. So, yeah, took a Waymo on the freeway there and back. Absolutely loved it. Got to say, I'm such a convert to Waymo or just self-driving for these purposes. But I know some people... I haven't done that yet, so I'll have to check it out. Yeah, it was still a strange moment, even though I've ridden Waymo a lot and still do, or like when I go to SF, but the moment it was pulling onto the freeway, I was like, this is quite weird, but then you get quite used to it. So, yeah. A little faster reaction time.

Slightly worse consequences if things go wrong. Yeah. Yeah, well, basically it just sits in the middle lane and doesn't go over the speed limit. That's what every driver should do quite frankly, rather than weaving around. But yeah, what else have you been up to in the last few weeks? Well, I've had a nice little break from work travel until today. I'm actually, I just got to Atlanta a couple hours ago for, I'm here for IBM Tech Exchange this week. So if I sound a little bit different, I'm using my travel mic. So that might be what's going on. But I have six talks this week, four on Tuesday. So hopefully I have a voice. It's probably good that we're doing this early in the week rather than later in the week, because I might not be able to speak by the end of the week. But I did do a fun solo trip with just my daughter, my four-year-old daughter, to visit my parents in Canada recently a couple weeks ago for Canadian Thanksgiving. So that was really fun. Nice. And then back, and then we have our big conference in Confluence version of what you did with Supabase that's coming up here next week. Yeah, of course. It's kind of crazy because even though Confluent is part of IBM, you actually still have a conference-specific conference.

Yeah, and then they're back-to-back, so that's fun. Yeah. I don't know how long that exists. Yeah, amazing. So, yeah, I mean, we've got lots that's happening, even though this is probably going to get released maybe three weeks since the last time we did this. But we've got a lot to get through. Start with the news, basically. Yeah, main headlines. And we're going to sort of roll in on, I think it was the last episode, we did a sort of A&R acquisitions and raises special, so to speak. But actually we're going to sort of roll in some more of that into our main headlines this week as well. There's a lot still going on, and it's good to keep up to date with the movements as well. but top two main headlines really the php vulnerability that seems to have affected 70 of the internet i found this yeah very interesting just the fact that this had been two bugs that had actually been discovered in 2024 already yeah i didn't realize that until i was kind of digging in more because i knew we were going to talk about this i didn't realize that it actually just been discovered you know several years ago but they just didn't consider it like

an exploitable thing at the time so they didn't prioritize patch. Turned out they were grossly mistaken, I guess. Yeah, well, I guess this is then the whole AI security story where OpenAI and Google in some AI form figured out how to exploit both of these. The biggest problem here was that it's just a complete remote code execution attack. It doesn't really need a user to do anything. It literally could just be you go and find a PHP-powered website and basically could take over it was the worst case of what could happen. So yeah, it basically meant that a lot of government sites across the world all had to be sort of taken offline while they patched. So there really wasn't any other option other than to just take things offline until fixed. Yeah, and like Wikipedia's PHP. The big thing is like the sheer number of WordPress websites out there. Yeah, still dumbfounds me like how much the internet runs on WordPress, basically runs on PHP. Fun fact, didn't know that PHP does sound for a personal homepage. Didn't know that. No, I didn't either.

I programmed in PHP for years, but I had no idea that the LAMP stack was so popular in that probably late aughts to mid aughts period, where there was just an explosion of companies and websites. And of course, popular CMSs like WordPress. And it's just kind of went everywhere. Nothing wrong with it, I guess. We've kind of moved on from the technology perspective in terms of what the modern web languages are now. but it's still everywhere, all over the internet. So I just point like that. Suddenly it becomes a huge, huge problem. Like Wiz described it, I think, as the most severe vulnerabilities in software history. Wow. Yeah. So yeah, it's probably anyone listening, probably you've been affected in some way in the sense that something you might have accessed over the last week or so was taken down for a day or so while this thing got patched. Apparently, though, it was the sort of PHP community and like sort of in the backroom channels they had actually been it was i don't think responsible disclosure is really a thing so much

these days with ai but at least people on the good side were aware of this and able to coordinate a bit more especially with the super important i think like government websites and and so forth before it was like really really known in the wild so that's a good thing for in terms of the community i still find this a bit shatter-fired just because i hate php as a language. There's a reason I tried it other than just put it down very, very fast, but I know that doesn't make it insecure. I just really, really... It's the only language I really just hate. It's kind of an odd language. I mean, I guess it has some... The fact that you're using dollar signs for variables is reminiscent of earlier web languages and things like that, but I think part of the reason why so many people ended up using it... One, you had the LAMP SACS, you had this open source version of MySQL, Linux, and PHP, where you could spin up websites. You didn't have to, you just didn't have as many options back then. Like, it was either sort of proprietary stuff that was fairly expensive,

or you could go with the LAMP stack. And then there was built up significant community around it. And even though the language has certain kind of weirdness where there's, like, a dozen different ways to do searches over arrays and things like that, it basically has, like, something built in to almost do everything. There's, like, some version of it. You might not be able to guess it because it doesn't follow sort of the standard naming conventions, and they use different naming conventions all over the place. But it's there. It's somewhere in the language. We'd call a language at this point. I think that's not exactly the case. But similarly with Python was never intended to be the language it is today. It was really supposed to be something like a wrapper for stuff that people were doing in like C++, and we're going to build my full applications in it. JavaScript certainly was never intended to be. I was going to say JavaScript. I mean, it blows my mind that JavaScript turned into the sort of dominant beast that it turns into. There'll be a lot of people probably listening going, well, JavaScript, like, what are you talking about? And I do, unfortunately, like JavaScript. So, yeah, I can agree to disagree on languages, as every good programmer should do at some point.

Great. So moving on to the other main sort of headline we've, way back, we talked about some espionage cases in tech. But this is a pretty wild one. OpenAI and an employee was discovered to have placed a sophisticated recording device behind a television in a conference room. And I still love the fact it was concealed in toilet paper. And of course, a lot of people have taken to Reddit saying like, oh, of course, I would never think to look at most TVs just have something in toilet paper behind them, don't they? It's like great concealing. Yeah, kind of a strange choice. Very strange choice. Yeah, maybe it was some device that needed some, I don't know, stopped two connections hitting each other. Who knows? And apparently it was picked up just during a routine sweep of conference rooms. So that's interesting that OpenAI even do this at all. I guess they should be on guard. I happened to go to, I think, this exact office a couple of weeks ago. And there's a lot of security. I think this was before it was known.

So it's a lot of interesting, a lot of security to get into that office in the first place. It doesn't have a name on the door. You have like three different things you have to do outside before you get inside. And it's already quite a lockdown place. And this Saudi national had placed this device in the back of the television. It probably was an executive conference room. But yeah, this raises a lot of interesting questions of like the why here. I mean, it does feel kind of state sponsored in a sense because this person apparently was a sort of go between Saudi Arabia and Silicon Valley. That was kind of how he had positioned himself, money in Saudi Arabia, opportunities in Silicon Valley, and now working for OpenAI. So it seems a very risky thing to kind of throw all away on this. But yeah, I guess this is the stakes that are there now with the Frontier Labs. Yeah, I just wonder what the plan was. Also, how did they track the device to the employee? Did you uncover that? They did a sweep. And then there's some screenshot of, like, showing the guy behind the television.

Yeah. I mean, there's probably cameras in there. I think it's just a camera, yeah. I think they managed to sort of hyper-zoom in and show this guy. I mean, they must have seen, once they had discovered the device, they must have then tracked who was going in and out of this room. And, yeah. Yeah. It definitely looked a bit sad. Who had a roll of toilet paper with them? Yes. Roll of toilet paper peering behind the television. It all looks pretty, yeah. I mean, it doesn't feel very sophisticated in total. So for such like high stakes, yeah. Yeah, it's very strange. And I think given, we've covered this a bunch, but given all the concerns over the Chinese labs and national security, so the arms race around AI, like does this point to like a potentially larger issue of where AI security needs to be some sort of national security matter now? Like if the US kind of wants to keep its lead and that's important for many reasons, does it need to be some sort of national concern at a government level? Yeah, absolutely. Well, clearly, I mean, given, yeah, Saudi Arabia, probably state-sponsored there.

So, yeah, it feels like perhaps it should become more of a national concern. And this is, I think, something that we sort of debated a bit either in the last SED news or the one before that. So, yeah, always something I'm sure we'll come back to. Moving on to the acquisitions and raises, A&R, so to speak. I know A&R is something in the music industry as well, but I thought I'd just reappropriate it for our purposes. So acquisitions and raises. The first one that has appeared in the last few weeks is Chroma, is joining Replit. So, I mean, again, I think we should always home in on the words being used because joining usually suggests acquihire. So, yeah, this does look like an alcohol hire that Chroma, which we had episodes on with just not that long ago, actually, with the CTO, Hamad. Hamad has actually left Chroma now and has gone to OpenAI. And I think it's his CEO, co-founder. He's, I think, some of the other team there joining Replit. One of the founders of Replit was apparently a key investor in Chroma.

So there's a link there clearly as well. On the tech slide, what's your take on Chroma as a product and technology and then what do you think Reflit could do with that Yeah I mean Chroma was I think they were one of the top vector databases especially when RAG and vector databases were really taking the world by storm And a lot of those companies raised pretty big rounds and grew really quickly, you know, a couple of years ago. But I think the challenge has been that vector search basically turned into a feature within many, many different databases. and as a result, I think people churn from these different platforms. Growth slowed down significantly. Vector search is still something that people often want to do when they're building AI systems, but there's many, many ways to do it. Even then, there's lots of ways where you might not even do vector search if you're doing search because you rely on other ways of searching and doing document lookups and just exploit the fact that there's a larger context when there's other ways of dynamically getting information

than just semantic search. So overall, the market kind of shrunk down on them, and I think this has been the consequence of it for many of them. I think it's easier for the existing databases to kind of pick up the feature than it was for some of the dedicated vector databases to really carve out market and continue to grow. Yeah, and Chrome as cloud offering will shut down with this. And, yes, it's sort of technology, not only open source, but kind of rolled into Replit, and the remaining team moving over there as well? I think this is, I think I mentioned this in a prior SCG news too, where I think that just because things are moving so quickly, there are a lot of these kind of moment in time, like products that will, in companies that will grow. They seem like great businesses for some period of time, but then they get essentially eaten up by other product areas. Like, I think it's very hard to have a company that kind of is the future at this point, because it either needs to grow into a platform, which some of them might,

or it ends up getting eaten up by the existing platforms. Yeah, and like sort of hyper example of growing into a platform, at least at the moment, is Vercel, who just seemed to be releasing product after product, clearly pivoting far away from the roots of where the front end to your back end is somewhere else. And basically they're just cobbling together all the bits that make a back end and, yeah, seem to be doing a great job at that at the moment. I think they probably saw where the market was going and there was less of a moat around their original set of capabilities. Definitely. Think about your mobile app source code. Once it hits the app store, it's out in the wild. And without the right protection, decompiling is easy for malicious actors looking to steal your IP or tamper with your software. That's where GuardSquare comes in. GuardSquare provides the highest level of mobile app security for Android and iOS applications and SDKs. Their advanced tools integrate seamlessly into your CICD pipeline.

We're talking polymorphic, multi-layered code hardening techniques and automated runtime application self-protection, paired with mobile application security testing and real-time threat monitoring, to deliver the highest level of mobile app security without compromise. Don't leave your hard work exposed. Secure your mobile applications today. Go to guardsquare.com to learn more. I'm a voice agent powered by Deepgram. I don't just understand words. I understand context. Like when you're fixing the sink and say, We've got a leak. You're talking about a pipe. But when you're a new parent and you say, We've got a leak. You mean that you're going to have to change into a clean shirt. Oh, it's everywhere. Deepgram understands words and context. across STT, TTS, and voice agents. That means better conversations at production scale. Learn more at deepgram.com slash context is everything. This portion of Software Engineering Daily is brought to you by Origin. Your coding agents can launch each other.

Theirs did. An engineer asked Codex to pull data from staging. Codex had no connection there, but CloudCode did, so Codex launched CloudCode. When CloudCode said it lacked permission, Codex relaunched it with the dangerously skipped permissions flag. Nobody was asked. A permission check doesn't mean much if another agent can switch it off. Your logs might show a process that started with a flag. Their trace showed the whole job and why Codex did it. Origin finds the agents on your company's endpoints and reconstructs their work from request to outcome on one timeline. Origin is endpoint AI observability. Read what they found in their own traces at originhq.com. That's O-R-I-G-I-N-H-Q.com. The next one is Dino, the runtime, so open source runtime being effectively acquired by Cloudflare. So this is definitely not something I think I saw coming exactly. Dino had been struggling for a while to kind of keep up the traction, I think, that it got initially.

But I don't think I would have predicted that Cloudflare would have vacuumed it up. This was super interesting. I thought originally that it was more of a talent-based acquisition, and I'm sure that's a piece of it. But when you read up on the articles around the acquisition, it seems like they're going to continue the project and continue to invest in it. So it seems like maybe Cloudflare was using it wildly. I don't really know. It's pretty hard to keep a language runtime as a successful solo company. That eventually gets a landing spot. either has to go some sort of open source foundation in company backers, or it needs to go and be part of some strategic reason for a company. Yeah, and it is going to have a working group partly set up. I'm aware of people in industry who have now been asked to sort of be part of that. Because obviously Dino has used just, it's been a huge part of the Superbase story, in fact, but also was actually a huge part of one of the areas of a Superbase product where a lot of feedback was, can we, a developer saying, I don't want to use Dino.

We were pivoting away from it already. So it's interesting. And then this news dropped as well. Then next one, slightly different sort of area, but a company, Sundial, has been acquired by Meta. And this is in the data and analytics space. Yeah, I didn't know a lot about Sundial, but it seems like they're planning to add that into their existing AI portfolio. But I think it's another situation where through Aquihire, the company's no longer going to be serving outside customers, essentially. So, you know, that Chrome is doing that with their cloud service and dials doing it with their customer, which actually includes OpenAI as well. So they'll lose access once the deal closes. So this isn't new just because of AI. There's always been risk with investing in a startup's product because you don't know the longevity of the company. I wonder, though, if that becomes even more of a risk now, just given the accelerated pace of technology. It seems like companies raise more money,

but they also seem to accelerate the pace of the product not being successful too. Yeah, absolutely. Definitely, Acquire, and anytime we see the word join being used, often indicates kind of that. The final one, yeah, is to do with Superbase, actually. it's a base acquired Terso. So this was a proper acquisition, acquiring the technology, but the team has also moved across the entire, I think there's about 10 of them, also remotely distributed teams, so quite an easy culture fit there. But yeah, Terso is basically databases for agents sort of from a technology standpoint. It's effectively an S3 file, so you don't need any compute to keep these things in storage and then it needs very little to then run them. So yeah, Superbase is going to be continuing the Terso cloud offering, but it's also going to be rolled into a new product called Spark. So obviously I can't talk too much about what that is,

but anything you can find online, which is already there. So this new product, Spark, is going to be a sort of incarnation of Terso. Terso were already, it's an interesting technology technology because they basically rewrote SQ Lite and Rust to achieve. So it's like an SQ Lite interface, but it is technically not SQ Lite. It's a rewrite, and they're doing the same with Postgres. So it's a rewrite of Postgres that they were working on. And that's obviously what Surface is quite interested in because it is a Postgres-first company. So kind of interesting. I'll not be able to say too much from the inside, but it is definitely interesting to see this acquisition happen. yeah between Postgres and SQ Lite you've kind of you got a good chunk of the market yeah it is interesting yeah certainly which parts of the market want Postgres or do they want SQ Lite or do they almost want both so yeah that remains to be seen but yeah very interesting times in that space and then yeah we've got a pretty chunky raise from a company that I had not heard of but definitely deserves some attention they're called Flapping Airplanes

So, of course, nice name that means absolutely nothing. But they just raised a Series A, a $1 billion of funding at a $6 billion valuation for Series A, which is pretty chunky. They're only one-year-old. I think their seed round was over $100, so. Yeah, seed round was over $100, $180 or something. But yeah, they're a one-year-old company, and they basically believe that GPUs are really not in any way being utilized to their full potential. And so they believe that they can train, especially training, they can do in a much more efficient way and utilize the GPU in a much more efficient way. So yeah, pretty interesting. It started by one of the founders also founded this sort of underground student-led accelerator called Prod. and there's only been a little bit about this written recently so this is all via tech crunch so thank you to tech crunch on this one but yeah basically prod is this sort of underground tech accelerator run by students at mit harvard and stanford cursor has is one of their graduation

companies but it's sort of you can't really apply to prod prod finds you sort of how it's built But yes, plopping airplanes is, I don't think it came exactly out of prod, but let's say that one of the founders is part of that. So clearly, yeah, I think investors are very interested in the people as much as the technology. Yeah, I mean, the idea of better GPU utilization makes a lot of sense. Like, there's a lot of waste right now. And I think essentially what they've developed is this idea of, like, a mega kernel that can run memory-bound inference alongside compute-bound training so that the GPU is rarely waiting. It's always being utilized in some way. So it's kind of like more of a less sexy than like AGI and some of the things that other AI companies kind of tout, but it's solving a real problem. It's more of a classic, I say, like computer optimization problem than we are typically hearing in the news and things like that now. So it seems like I didn't know a lot about the company either, but it sounds like, you know,

what they're doing makes a lot of sense. It's probably hard problems to solve there. So if you can solve them, then it kind of justifies the raise and the valuation. Yeah, well, I know a billion dollars can go pretty fast these days. Especially where TV is. Exactly. But still, you'd expect something to come of this, so we'll be following along. If you check out their X announcement, I think on the, I don't know if it was about the Razor, just about the product in general, but they have a little video. Of course, Doom is part of it, which is great. So it's basically a video showing that they are trading data and serving, and it is playing Doom all in one kernel on a B200. So that's kind of fun to watch that for anyone interested. I love that they still reach for Doom, despite the crazy advances of technology there. So nice to see. On to our main topic. So we're kind of looking at ways that AI products have died. or you could call this the AI graveyard

it was inspired a bit, there was an article in TechCrunch about this and looked at some of those examples but very much across that spectrum and we're going to look through four ways that we've seen AI products effectively die over the last year or two given the speed at which products have come about and then they have fallen away and both products that were just that was the product of the singular company that did it, or products that have been within other products that you might know of. So like Notion is a good example. They have various product offerings and made big splashes with some of them, which we'll talk about, but actually they get folded pretty fast, which is what people don't talk about so much. So we're going to go through kind of some of these and see how these happened and like what that maybe means for just the AI product landscape in general over the next few months. So we have kind of this theme of the platform ate it. So effectively a standalone product whose feature became a sort of checkbox in somebody else's app.

So the first one in there is Relay. That was actually launched in 2021. It was an AI kind of Zapier alternative. I think they shut down because Google and OpenAI and others basically built the same feature into their products. Yeah. They didn't have much of a competitive mode, essentially. It's not that hard to recreate what they had. Yeah exactly So yeah when we talk about a platform kind of eating it it like the large platforms basically just replicated it and really just shut down Yeah I mean this goes back to what I was saying earlier about like even the vector databases and I think a lot of other things like these kind of these moment-in-time product offerings where they can't really defend the space that they have from the platforms because they're like a singular feature, and it's either not a complex enough problem that's not that difficult for another company is motivated to recreate or it makes a lot more sense for it to be part of kind of a larger platform because more and more, especially in the enterprise,

companies just don't want to have to deal with a million different vendors. Yeah. Another one is Hux, H-U-X-E. This was built by a former Notebook LM developer, and it was sort of intended for podcast-style audio from your content, which is kind of how notebook lm got its big break out of google anyway i still remember this like yeah two two and a half years ago everyone was suddenly creating these ai podcasts of their content and throwing them around because they thought that was an easy way to digest information interesting that then i guess one of the people left google to then kind of just found a version of this on its own but then platforms like spotify actually just shipped a similar feature within the platform and so having a sort of dedicated platform to this just didn't have a lot of legs. I was kind of surprised that a company thought it could just do this or it just didn't spend enough time then building itself out into a sort of wider platform. Yeah I mean even the audio

feature has been built into one of the chat based systems like ChatGPT and things like that too so it just ends up being like a small feature in the UI of some existing product. Yeah and then And finally, in this sort of the platform, ate it as such, ChatGPT Atlas. I have to look this one back up again. It was actually a browser, basically. I think we talked about it. And I think we were not, if memory recalls, I think we were kind of not bullish on it. Yeah. I like, though, that they keep the blog posts around. Like they kind of say this product has been deprecated, but here is the blog post that we put out about it. That's kind of fun that you can, I mean, you can find it on Wayback Machine, but just kind of find it right there. so you can see what they were trying to achieve with it. And this is only back in October last year, so I guess a year. But they did shut it down in August, and some of the features got folded into a desktop app and Chrome extension. But yeah, the whole browser play perplexity went big on that as well.

But I really feel like the whole browser play has just kind of disappeared almost. Yeah, I think that disappeared. And I also think, kind of going back to what I was saying, and the accelerated pace where these companies can kind of learn that something's not successful. You think of what is the bar for success that a company like OpenAI is pretty high. So if something isn't like hyperscaling in the next two quarters, then it's probably potentially going to see a swift step because they'd rather spend their resources on things that are going to be those hyperscale opportunities. Yeah, and we work with them quite a bit. And again, can't say too much, but I can definitely attest to the fact that products do sort of appear and then they disappear quite fast. Even nothing's particularly built. Hey, we've got this idea. And then that kind of don't hear about it for a while. So they've had like nine different agent frameworks or takes on agents. Wow. They keep trying. They keep iterating and things like that to figure out. And they have the scale at the moment to do that.

they have the scale and the money it seems to just kind of really throw stuff at a wall and see what sticks yeah it's like the google labs era of google like on steroids yeah yeah yeah so yeah very kind of interesting there so then there's a sort of another next category is basically that so you can subtitle it like you can't afford your own success i.e you literally cannot afford to run the thing that is a runaway success in theory but just too expensive to run or there's an example here where it's both expensive and ends up with a bunch of legal liabilities so the first one is Sora people oh yeah Sora I remember that that was the video model from OpenAI and they even got to the point where they had like a deal with Disney who were going to like license their kind of I content into it. It wasn't very formed. I don't think that it just had a 1 billion investment from Disney that would include licensing. But yeah, they basically just couldn't sustain

how much money it was costing them to run this for users. Apparently, one user managed to rack up a million in one day, like a million dollars of usage in a day. I think there's even a $15 million in a day reported as well. So yeah, they just couldn't sustain what this was actually going to look like commercially and sort of similarly character.ai shutting down basically partly to do with how much it was burning like i think it was burning sort of eight to nine million dollars a month but it then had a ton of legal liabilities from unfortunately people saying that it did people talking to it came back with answers that were quite unhealthy unfortunately for a lot of people and yeah just ends up in chapter 11 i believe on october 12th so that just shows if you do have a you know you want to create the next ai chat experience or or whichever like it is very very clear now that safety is a is a big a big concern that that company should have it was also like a lot i think it was a lot of miners that used

yes exactly and then yeah the controversy around it was pretty if you read into it it's pretty awful and pretty scary. Exactly. Yeah. So I do see that as a, quote, good riddance, but it's interesting to call out just sort of how, yeah, how a company that just doesn't take its responsibilities seriously enough, yeah, just ends up getting legally blitzed. I still think that outside of engineering, of course, it's still trying to kind of find like what is the killer app. There's the chat experience, which is, you know, of chat GPT is obviously like a kind of the killer experience. And then sort of general purpose, like question, answer, exploration. And then there's agentic engineering. But a lot of these other attempts for things like create a, especially in consumer, like create some sort of social thing that I can interact with and things like that. But yeah, I just haven't quite found what that killer experience is. Like there's for mobile, there was Instagram and Uber that came along that were these like mobile first companies. And it took several years of mobile before they were able to think from sort of a mobile

first lens and do something that wouldn't have been possible if people didn't have computers running around in their pockets. But I think we're still struggling to find that with AI. Yeah. Another subheading topic is agents effectively killed the interface. And I think there are going to be many, many more of these in the future as well. But two that stand out, Notion Mail, the first one. So they launched that in April 2025, but they shut it down, I believe in June this year, or they announced they were going to shut it down in June this year and then closed it completely on September 22nd. And they basically said, Notion said that more than half of its users just ended up managing email without even opening the inbox because agents did it for them. So that's kind of interesting how, yeah, they thought they could come up with this interface that was going to be so much better than Gmail. We've seen that story before, Superhuman, which also, Cyborg, you know, got acquired by Grammarly. Grammarly and I'm not really sure

how much, I think the superhuman product still exists but I just, I don't think it's probably getting a ton of active development on it. But yeah, it was a bit of an odd one for Notion to go hard into this because I guess they were sort of matching AI with email interface. I also ran a startup that was kind of in this space but quite quickly we saw that having an agent work against the inbox was probably going to be where things were going but it was also just an incredibly competitive space to be honest. So it's clearly, it is interesting that Notion also found it competitive to the point that they thought that keeping that product up wasn't in their best interests. Yeah, I think that agents that only exist in whatever their system of record is, is kind of something, it was popular like a year ago, but it removes a lot of the value of the agent itself because you're limiting its capabilities. I think one of the reasons why Cloud Code and other agents that kind of run from your terminal have been so successful and so popular because you kind of point them to multiple disparate data sources,

and then they're able to pull in data from those different places and reason about basically connect ideas together that would have been really hard to do previously without wiring a bunch of APIs together into a dashboard or pulling data and centralizing it somewhere and then trying to pull a dashboard together or something like that. So you're able to federate that out. But if the agent exists in the system of record, you can't do the federation piece. Its knowledge essentially is always going to be capped to buy whatever that data exists in that system. Yeah, and interestingly, Resend have actually spun up a new product in the last couple of months, or at least they've launched it in the last couple of months. Yeah, it's just literally called Inboxes. It literally is the inbox for agents. I mean, I look at this and kudos to them. They have a really, really nice landing page for it, I've got to say. But I'm still a little bit confused on the use case. The use case that they seem to present is sort of killing the front, which is like a customer support, or a Zendesk. So it's sort of like the examples they show,

you can basically build your own kind of Kanban from inbound email, i.e. people asking you to do stuff. Yeah, or you can also have people emailing in for jobs or something. But yeah, I don't feel their use cases are quite explained yet. I would still support that there's a case for email protocol being used by agents because a lot of it is very clunky to use on an interface. And like, yeah, you could obviously bring up the federated piece, Sean. But yeah, I'm curious to see how this Resend Inbox for Agents pans out. Maybe it'll be the feature of next year's AI products that died. But great to see still Resend trying other things than just being a sending protocol. So, sorry, not a protocol, but a sending platform that they do very well. And then we've also got in this kind of bucket we've got, which is, you know, agents taking over interfaces. There was a company, Yup, Y-U-P-P,

which was a sort of a model variant of this. It was 33 million A16Z crypto-backed playground for comparing 800 models, but shut in March 2026. and the founders said that models had improved so fast that the need for human comparison had faded. So yeah, that's an interesting one where you think you're kind of in the right space at the right time, but actually things, as we've seen over the last couple of years, do move at such a pace that it's very hard to actually keep up with what is relevant for a developer to need. I wonder what the ultimate business plan was there. Surely it couldn't have been just, hey, our business is going to be model comparison. I don't really see how you would turn that into a business. There must have been something they'd planned to go beyond that. But I'm curious about what that was and what the plan was and why it didn't work. Yeah, I mean, if I go into their, they've got a site up who's kind of saying what they did.

And obviously that it's all shut down now. But they said provided free access to the world's most powerful AI models. Over a million users across 200 countries participated. but yeah it doesn't really sort of answer that question having users participate to evaluate and give that sort of human evaluation it feels like almost it was traced a bit of a weird maybe trace data kind of play where like people would if I look at it actually that's a little bit of what it looks like it looks like it's a sort of gateway that they might take the trace data and then maybe somehow monetize that I don't know like from from the human evaluators exactly yeah and like what we touched on in a slightly different form but like what we touched on a couple of weeks ago, a couple of episodes back on the Chinese proxies. Yeah. So I don't know, maybe they got into hot water there because it kind of was a proxy-ish thing. But yeah, it wasn't clear what the end goal, business plan, any of that. Yeah. And then finally, we've got sort of a topic here on like AI that sort of failed,

but this is also AI that has never really materialized. so yeah we've got a couple of actually hardware ones we had the humane ai pin which raised 230 million former scd episode as well yeah did you do that one shot yes i did yeah all right okay so yeah but they unfortunately their assets apparently were sold to keelopackard for 116 million in february of last year so yeah i mean we'll get into you spoke to what's maybe your take on that It was an ambitious project. A lot of smart ex-Apple people went there to try to do something big, sort of change the world. And, of course, doing anything big, like, comes with risks. I think it really hard to do something where you trying to rethink what is the fundamental interface that someone going to interact with technology And that kind of what they were doing Like it didn depend on your phone It was a closed system So you not getting any benefit off of the existing like phone ecosystem It had no UI. It used, like, a laser.

Like, they're trying to rethink everything from bottoms up to have, like, a non-invasive piece of technology that, you know, serves you when needed. There was other problems, too. like their launch was not very successful like the initial launch because it just didn't work as advertised the demo looked great but the actual reality of the device was not very functional it just took way too long to do compute models were much slower they could only put so much stuff on the device and you have to round trip across the network your network connection's not very good you're like five minutes before response it's like you know there's a lot of a lot of problems to try to navigate with a lot of these hardware devices that you're supposed to be using live in the wild, what variables to manage. Yeah, and like a sort of similar, it's still going, but Rabbits, so the Rabbit R1, which was, as I say, was, is this product, this orange box thing with a bunch of buttons on it where you can do all sorts of AI things,

it's sort of meant to be your sort of AI companion. But yeah, it really got pretty flat reviews. I remember watching a Marcus Brownlee review of it who just didn't seem to like it. He did sort of review it as quite unfinished as a product. Rabbit then said they were building a thing called a Cyber Deck, which some people sort of then said, well, it's basically a netbook, but let's see it. But actually, we've not seen it. That hasn't materialized. And actually, if you go to Rabbit's website now, they're kind of really playing on their own agent OS. So actually maybe more of a software play, which probably would be more economical, it sounds like, if they're not really hitting it out of the park on anything hardware-related. So yeah, or just trying to monetize basically what they maybe have on device on the R1, I don't know. But yeah, definitely a bit of a muddled company, even if they seem to be trying very hard on certain things. Like, hardware is hard, as they say. I think hardware and voice is tough. Even if you look at Alexa and the Google Home devices,

they've been largely given up on. They tried to turn them into app platforms. It never really took off. For whatever reason, that form factor. I think one is it's hard, so the performance a lot of times isn't very good, and then people get frustrated. And then I think we made this comment before. They end up turning onto just fancy devices to turn the lights on and off and set timers. Yeah. Yeah, exactly. And then just to round out this one, I mean, we've touched on certainly one of these before, but like Apple's Siri AI just delayed repeatedly. And actually, there was a, we touched on this before, but there was a legal settlement on iPhone 16s because they claimed they were going to ship with this amazing Siri AI, which wasn't really the case. It was all a bit fake on stage. and there was also a product called Microsoft Recall. So I don't know if Recall's even, has it actually launched in the end or? There was a ton of controversy around it because the idea was it takes like periodic screenshots so that you can search your history.

So I don't know what the timing on it is, but it's like as I'm working or as I'm using my computer, it's taking screenshots of what I'm doing and then I can search that history. It's almost like your browser history, but it's your entire desktop and then visual. So people are like up in arms about the idea of that. So, I mean, it comes down to like, you know, sometimes with some of these products, and I think we brought this up before when we talked about some of the stuff that like Meta has come up with. And it's just like, who's this for? Who was asking for this? What problem was that person trying to solve where they thought this was like, oh, you know, what would really help me is if I had something automatically take a screenshot of my desktop every 30 seconds or something. It's funny. what's interesting there is there i think it's it's toby lukke he's built his own version of this like a while ago because he says he has this like crazy library of every 30 seconds i think for him or every minute he's got a screenshot from like the year 2000 since right and like for him i can imagine that's insanely helpful where being on the forefront of having access to ai models when they

came about and so on, he's been able to turn that data into something quite interesting. But yeah, the average person, and especially when it's a company like Microsoft, pushing this on consumers and saying, hey, guess what? We're going to take screenshots of your screen every... Yeah, I can see why they didn't get a great reception. So I'm just like, who's it for? Well, it clearly might be for an esoteric CEO of Shopify, but it's not for the... It's a small... It's a very small subset. Total addressable market, though. Yeah, so I'm curious if they were, like, inspired by that and then just didn't do their kind of market research. So, yeah, very interesting. Yeah, it's supposed to be in Windows 11. Okay. Periodically take screenshots to your screen to create a searchable visual timeline in your past activity. Yeah, but then there's also now certain Brave browser and AdGuard saying that they've added functionality to basically counteract this. great well yeah that's a very interesting spin through and for our main topic of yeah ai that

failed effectively but you know failure is part of the path of making technology at all yeah the other thing too is like every cycle has a graveyard there's lots of famous.com you know pets.com is like the classic one that people always refer to that lots of companies didn't make it there the majority companies that are started never go anywhere and most of them you've never heard of because they never went anywhere. Most companies fail. It's a fun spin, but it's not like it's an AI-specific thing that people should keep you in mind. Yeah, that's a good point. We've focused on AI today, I guess, given that most companies in the last two years are pretty much AI-focused in some respect, even if it's just in name to make sure that investors are interested. But it's definitely not an AI problem per se. It is part of the landscape that comes with technology and venture investing and startups in general. So yeah, moving to the final piece of the show, Hacker News highlights. What did you turn up in the last week or so?

Yeah, so there was a fun one that caught my attention, which was that washing machines took decades to catch on, which is kind of a bit different than the typical ones that I highlight, but I thought it was interesting. Essentially, apparently, and I had no idea of this history, but like washing machines came out in like the 1910s, but it took until the 1950s to win over American homes. And even though like laundry is one, was like the most hated household chore. And I still, I personally can't stand. I don't actually mind putting the laundry in the laundry machine. I don't like the folding and organizing afterwards. But so there was a couple of different reasons for this. One was that the early machines still needed someone to feed the clothes through like ringers by hand. So it saved you effort, but not, it was still like very time consuming. And then there was also this social status issue where in the South, many white families kept black laundresses to do the heavy manual work because they felt like it was beneath them at that time in the era, the U.S.

So there was that factor as well, where it wasn't like a problem for these people who could afford to have people basically do it for them. And then it really wasn't until after World War II that they became much more common. and there was laundromats that were built up as well for people that lived in places that were small on space that needed somewhere to wash their clothes and that's kind of what ended up accelerating the sales of the washing machine but it took like almost 50 years for them to take off yeah it's super interesting yeah i always find any of that kind of like domestic history like complete sidebar but like the fact because living in asia and you know a lot of people eat out as supposed to cook in their houses and then like in europe it's kind of the opposite but there was a time in europe when eating out was the norm because people didn't have but actually then having a kitchen in your home was seen as like very upper class so then everybody wanted a kitchen in their home whereas i think we would often see eating out as a more expensive and trendy thing to do the laundromat version of this and i guess a world war ii and just the sort of booming domestic

appliances being bought by these families that like the men had come back from the war and like and they were getting jobs and making money and i'm obviously just pulling on my mad men knowledge here watching mad men so much but yeah that's it's very interesting so yeah on my end there was one that i just kind of went by points to see i was just scanning around and i saw that this is when i pulled it yesterday it had over 1500 upvotes points sent in by user So Cam Young. Yeah, this is great. It's a local Wikipedia. So this guy, Ed Jefferson, edjefferson.com. If you go to edjefferson.com forward slash local wiki, basically you do need to enable it to find your location from the browser. I think maybe you can enter it manually, but it just creates like a very normal looking Wikipedia front page, but completely tailored to like a certain radius that you are sitting at at that point in time so yeah it's pretty cool

because like i was just sitting in a cafe and went to it and then suddenly had like a wikipedia front page with all the kind of landmarks and just like interesting houses and buildings that people have written articles about so it's like a more interesting kind of local travel guide i guess because you just never know what people have written about from especially from like a buildings and history perspective. I think it's awesome. I'm going to use this a lot more just if I'm somewhere and check it out. Yeah, that's cool. I'll have to check it out. The last one I had was this one. It was called Bring Back the Cool Save File Screen. So the author, I think his name was Aaron, but it was posted by Surprise Talk on Hacker News, was looking back at the save file screens from the sixth console generation. So Dreamcast, GameCube, PS2, Xbox, because most of those consoles at the time needed a separate memory card. each one had a menu that saved or saves as these small icons drawn from the game itself. So Dreamcast used this little portrait of the squares over a cloudy background.

Xbox had this kind of alien-looking green menu. And then modern consoles, they just show the game's cover art, which is kind of fine, but they're not that interesting. The author, this is like a plea to basically bring back the old screens, have some fun with it, and do something unique. that's interesting i'm yeah i'm looking because yeah ps2 i remember i remember very vividly the ps1 which was very 2d and then ps2 had these yeah 3d kind of icons i'm just looking at the one for ps2 at the moment so yeah like tony hawk's pro skater is like a little kind of bubble thing that just says tony hawk's on it and but yeah like memory cards i mean if anyone after a certain point in the time nobody will know what these are but yeah just literally to save on your console you had to have a memory card. And if you ran out of space, you had to buy another memory card. And there's only two slots as well. And then I think maybe PS3 enabled you to plug an external hard drive in or something as well. But then by that point, they had memory. So clearly I get excited about this, so I'm going off on a tangent on that one.

Finally, my one, just a kind of funny, quirky one. This was posted by someone called J-L-L-Y Hill. it was a very sort of well yeah a very detailed article about a bug in Figma that wasn't really a bug and it's basically if you pressed shift plus the number 2 on a Windows keyboard if you used the number 2 at the top of the keyboard you would zoom it would do what you expected whereas if you did it on the numpad it would actually act as a down arrow and people were like this is a bug why is this turns out it was actually very Windows specific that the way that Windows keyboards evolved from the IBM days was to do with NumLock. And NumLock was this thing that meant that if you wanted to use the numpad as arrows, you could do that. And this was like in the era of spreadsheet software coming into play. But then you didn't really know if NumLock was on or not. So then they developed this new thing. Mac never had this. Mac just basically had arrow keys and number keys.

So it was like a bug that goes all the way back to those days. And another fun fact was the guy, David Bradley, didn't write the article, but was mentioned, who invented NumLock, also invented Ctrl-Alt-Delete. And the person who wrote this said, like, that's a story for another day. So I hope that we maybe get that story sometime. Like, where did Ctrl-Alt-Delete come from? Yeah, that's interesting. I wonder where, what the origin, I'm sure we can look it up, but what's the origin of Ctrl-Alt-Delete? Yeah. Well, yeah, that's it for another SED news. Yeah, I guess we'll be back in another month taking a spin around the headlines and a topic and Hacker News again. But thanks for tuning in. Yeah, thanks, everyone. Cheers.