 Daily Tech News Show is made possible by you listening right now. We could not do it without all of you, including Paul Boyer, Brad, Kevin Morgan, and our new patron, and everybody welcome at it. On this episode of DTNS, Dr. Nikki explains how stingrays are making us blue without the chemicals. Plus, are you okay with AI generated images of food in your restaurant menu and rest in peace, apple car. We barely need. I mean, we never. This is the Daily Tech News for Tuesday, February 27th, 2024 in Los Angeles. I'm Tom Merritt and from Studio Animal House. I'm Sarah Lane from Lucidon in Alabama. I'm Dr. Nikki Ackerman's and I'm the show's producer, Roger Chang. Oh, my folks, so much fun news about apple killing products. Usually it's Google kill in a product, but today we shook it up and apples killing one that never actually was a product. So I guess they're not really killing a product. They're killing a dream. They are killing a dream. We'll get to that, sadly, after the quick hits. Sony announced it's laying off 900 employees, about eight percent of its global headcount and closing down its PlayStation London studio. This will impact a number of its PlayStation studios as well, including Insomniac Games, Naughty Dog, Gorilla Games and Fire Sprite. Sony also just missed a PS5 sales target, which saw Sony stock lose ten million dollars. Analysts also pointed toward a near decade low games margin, meaning the cost of making games is eating into those profits. Universal Music Group and TikTok did not renew a second license. They didn't renew a license for UMG distributed music at the end of January, and that's when around three million songs disappeared from TikTok as a result. A separate deal covered songs published by UMG writers. So that earlier one was for the recordings. This is for the writing. That deal ends at the end of February at the end of this week. And if a new agreement isn't reached, that means four million more songs would not be available on TikTok. But TikTok's just jumping the gun and muting them anyway, starting now. They've just been going around saying, well, these are part of that deal. So we're going to mute them. Since it's based on a songwriter, it means it won't be consistent by artist. So if you're looking at one of your favorite artists, three or four of the songs might be muted, but not all of them. Some songs from an artist will remain up if they don't have any contributions from a UMG songwriter. Netflix used to let you subscribe to the Apple app, meaning it had to give Apple a cut of its feats. Then in 2018, Netflix stopped letting new subscribers sign up that way, but still let folks who had signed up before through Apple keep being billed that way. Now Netflix is changing that too. Users who were being billed through Apple will now have to provide Netflix another method of payment or lose their subscription. It also means they lose any lower prices that they had been grandfathered into as well. Yeah, and you can't use your Apple gift cards for it either. The Humane AI pin, that's that little square. Remember they did a TED talk. You wear it on your shirt, voice activated, projects onto your hand for settings. It's going to start shipping out in March in the US. Probably start arriving to customers mailboxes in mid-April as a reminder. It's $699 and $24 a month because it gives you data service. You get a phone number, unlimited data, as well as the ability to ask any question and get an unlimited number of answers from their large language model. Human or Humane, I should say, partnered with T-Mobile for data service in the US and just announced that South Korea will be its next market in partnership with SK Telecom there. And SK Telecom is going to license Humane's Cosmo-S. That's too clever. You can't say it out loud. It's the that's their operating system. Cosmo for use in its AI products. SK Telecom announced a partnership with Perplexity, an AI search engine on Monday. So they're beefing up their AI repertoire. Nvidia's GeForce Now has a free tier that lets you queue up to get assigned a server to play an hour of games for free at a time. Starting on March 5th, Nvidia will show players waiting in the queue up to two minutes of video ads. There won't be any ads during the gameplay itself, however, but you'll have to wait through the first. Nvidia says it hopes the increased revenue will reduce to average wait times for free users over time. Yeah, because they couldn't add more servers or people just get frustrated and leave the queue in one of the tier. Everybody loves ads, Tom. What are you talking about? Yeah, yeah. All right, let's talk about this Apple car story. RIP Apple Car. Bloomberg's Mark Garmin reports that Apple has officially canceled Project Titan, a.k.a. its electric car effort. Apple CEO Ted Jeff Williams, rather, and Kevin Lynch had been leading the Apple Car Project since 2021, apparently surprising employees with this internal news on Tuesday to the more than 2000 people working on the Apple Car Project, some in software, some in hardware. At least some of those employees are said to be moving to Apple's AI team, which is led by John John Andrea. But German sources say there will also be layoffs. Now, rumors of Apple's forthcoming electric car have been around since 2014, including everything from a fully autonomous vehicle to a scaled back software vision, most recent whispering of a 2028 release date for something. Apparently it is not to be. Yeah, I think a lot of the knee-jerk reaction to this is going to be like, I told you so Apple could never do a car. It was never real, blah, blah, blah, blah. And that's fine. You know, if that was what you've been saying, take your your I'm right flowers and enjoy them. I think a more enjoyable, more interesting and useful way to look at this is Apple is big enough with enough cash that this is what a like R&D project to see if something is worth doing looks like at Apple scale. You can do it for 10 years. You can have 2000 employees and then you can say, you know what? We searched every avenue. We tried everything. It's just not going to work for us. And then you end the project. That's actually, I mean, at this scale, it seems a little irresponsible because of the cost and the effort that went into it, but it is the right way to do it. If you're like, yeah, this isn't going to work. Let's let's not keep, you know, throwing bad money after good. I mean, what would be more irresponsible is pushing out some sort of a vehicle that isn't ready in the last, I don't know, year or so. There's all sorts of, you know, conversations about like, are we ready? Yeah, are we ready for autonomous vehicles? Are, you know, are we ready for this? You know, what, you know, and the market is saturated. I mean, there's all sorts of reasons that I feel like it makes sense that that this got scrapped. I would have liked an Apple car personally, of course. I also felt like it made more sense for as cars, you know, increasingly, you know, the dash could be, I don't know, any sort of operating system type thing. That made more sense to me. You know, I would like to get in my Volvo, turn it on. And it's, you know, it's all Apple, you know, across the board. You know, you know, make car play better somehow. I don't know. But yeah, I don't know. Nicky, what do you think? Part of me is like sad for the 2000 employees. And part of me thinks it's kind of funny that they were like on this super top secret project for years and years. And then they couldn't tell anyone about it. And then it turns out it's not real, but kind of was certain kind of in 10 years. That's a lot of time to work on. It was real, but they couldn't tell anyone about it. And now it's not happening. And now it's not going to come to market, which is, yeah, I have a friend who just got laid off from a project that she's like, yeah, and now that project's never going to happen because we all got laid off. God, that's terrible. But but yeah, my first thought is like if they came out with a car, yeah, it would be the same feedback that's happening with the Teslas. And even Teslas are like proven and they're still running into errors. There would be this whole like, oh, again, Apple make a car. Aren't they a computer company? What's going on? And well, the amount of regulation that comes with anything like this. You know, any time there's, you know, an autonomous vehicle, I really do think that Apple, especially because they're retaining some employees anyway to work on, you know, AI efforts, all of that still can go into something that we know as car play now. And yeah, they can do something very big later. Yeah. Yeah, I think that's a really good point is the other thing that Apple gets out of this is again, it's simplistic to look at it as oh, they threw a bunch of money after the stupid dream of making a car. And then they failed. They developed a lot of technology in pursuit of this. They decided that this technology didn't meet their standards of the kind of product that they would want to get. Sounds like from the things we've gleaned over the years that they tried several different approaches from building an entire car to just making software systems. So none of those worked to their satisfaction, but what they learned and developed is going to show up. Keep your eye out for the story that says like, oh, and this new feature of whatever Apple product was developed originally as part of project type. Yeah, exactly. And I mentioned early, it's probably because they just found out that you're not allowed to drive while wearing Apple vision pros. So they just cancel the car. Yeah, it might and it might be as simple as that. That's like, wait, what? But, Nicky, you can see when you're wearing the vision pro, you can see. Still a bad idea. I still probably a bad idea. Or they just decided that to make all the software they developed for project type in a simulator for the Apple vision pro. It's actually just going to be a simulated car. There you go. Yeah, you can pretend to address. It's a racing game. Yeah, exactly. Come on, everybody. You didn't know the project's not dead. It's just Apple. Let's talk about delivery apps. Favaro for Media's Immanuel Mayberg has a report on ghost kitchens using generated images in their menus. If you don't know what ghost kitchen is, it's a kitchen run by ghosts. No, it's actually a term for. Yeah, it's a haunted. It's not a haunted restaurant. It's a restaurant that doesn't have a location you visit yourself. So sometimes they just rent out warehouse space. Sometimes it's somebody's house. Other times it's an actual restaurant like Tony's Pizza, but they're making food under another name like, you know, Mexican specialties or something. Jerry's Pizza. Yeah, but they show up on the delivery app just like other restaurants. So you don't know that they're a ghost kitchen. Many of them, though, it's unclear how many have been using generators like Stable Diffusion and Dolly to make the images to go with their menus. 404 contacted Reality Defender, which is a deep fake detection firm and verified to Mayberg that many of the images appear to be generated by a diffusion model. Grubhub specifically told Mayberg it forbids that. And if it will catch them, it will remove them. DoorDash doesn't have a policy against AI generated images, but says the images must be realistic depictions of what you'll get. And it encourages restaurants to take photos themselves and even offers complementary photographers, although one restaurant tour didn't like their photography. But, you know, DoorDash is giving you a free photo session if you want it. So why would a restaurant decide to use an AI that's going to show a shrimp with three tails or a fork with six spines or whatever? There's a few reasons. First of all, it's hard to make food look good in a picture. That's why commercials spend a lot of money on photo shoots, you know, spraying shellac on the food and everything to make it look unrealistically good. Most restaurants don't have a budget for that kind of thing. So AI is a cheap way to do it. Another thing is that the algorithm rewards refreshing your photos regularly. So again, that's that's time intensive, if not costly, unless you're using generated images, in which case it's practical to swap out the images a lot and boost your rankings. And the other reason is consumers respond to good looking photos. We know those photos aren't necessarily what the food will look like when we get it, but it's just an involuntary part of our brains that's like, oh, that looks good. Makes you want to order from that particular place. I mean, yeah, tell that to all the Instagram influencers taking photos of their food, you know, yeah. Since Instagram launched in 2010 that you can't make food look good. Yeah, you can. Oh, you can. It's just it's just hard. I mean, the influencers will tell you it's not easy. It's you know, you've got to take some time on a chair. You take a picture of your soup. I'll eat it. But no, I get it. I get it. We've got, you know, for the video folks, we have some photographs that were human made versus AI generated photographs. I think the human photographs are fine. I also feel like when I'm ordering food, I expect it, you know, it's going to be packed up in some way, you know, if it's sushi, it's not going to be like the restaurant. Yeah, you know, it's not going to be on that nice wooden board. Yeah, you know, you're you're you're you're phoning it in a little bit, you know, you're also, you know, paying for the convenience of not leaving the house, whatever it is, I'm not looking for perfection here. I also feel like if you have an AI generated image of something that's wildly different than what I receive, you know, maybe I, you know, open it and go like, are you kidding? That's not even what I ordered. There's no noodles in this noodle dish. Yeah. Then it's like, OK, I mean, maybe you fool me once, but then I'm never going to order from that vendor again. So it's like, it doesn't really pay off. Yeah, I think for this, it's the same problem that we always have with AI is that on the surface, it seems a little bit harmless. But if it just goes unchecked, it can lead to some problems. Like you could just have like a complete ghost restaurant where none of the food is real and they just serve you slop and you end up paying for slop and then you can't like go back because they don't exist. I don't know. That's like an extreme example. That restaurant would not last very long though. True, but they they could like pop up and get a bunch of people and then disappear and I don't know. This is my break. It's like a mudding, londering scale. Well, it's not unlike a lot of the stuff that you see on again, Instagram, you know, stuff that or even Amazon. Exactly. Stuff that's for sale where it's like it's cheap. You know, they just want to get some merch out. And once people don't like it anymore, they just change their name. Yeah, I this could lead down to some unsavory ways. I feel like I just want AI images to have a little watermark on them and say this is a generated and it's OK if that's how your noodles look like. But if you make something look really amazing and it's not, then it's it's false advertising and it's not very ethical. Yeah, part of me is like, well, I'm going to ignore the photos most of the time anyway, because I know that, you know, that I will end up being disappointed if I make my decision on the photos. But at the same time, I just I don't think it's right for the and that's why I'm glad that Grubhub and Dordesh have policies that encourage the restaurants to take actual pictures of their food because that that's what I would prefer. I would prefer if we were not, you know, using pictures that aren't even the actual food. It's one thing to glue the mozzarella to the pizza, as somebody was just saying in our chat, you know, to make it look like it sticks better. It's another thing to like have the pizza just made up a whole cloth by Dolly. And it's like, I don't even know if that's anywhere close to what their pizzas. In this case, you might as well use a stock image like why don't they do that? It's the same thing. And remember, it's not just us. And remember, it's like grandpa ordering off of Grubhub, who doesn't know any better and then who might spend extra money because this looks so amazing and then gets ripped off. Like I feel like it's not very fair. Well, yeah, I mean, it's it's akin to me being like, I want ramen tonight. In my mind, the ramen looks like this. It's like, I mean, show me whatever image you want. But if that's not what I get, then, yeah, you know, I will I won't be a repeat customer. It just makes me suspicious. I'm just like, if you're doing this, I'd probably just straight up won't order from you. And I like to tell myself I can see what the AI looks like, but maybe I can't. The ones that you're showing up on the video are a little harder and harder. Yeah. Yeah, that's true. And you'll just stop. You'll stop believing all the photos if all of if a lot of them are AI. Because you're like, I just assume. Yeah. So it's just not good policy. It's a note about the algorithm rewarding an establishment, you know, pizza place, whatever, for, you know, changing the photos out all the time. It's like, well, OK, I mean, I guess it, I guess. But what if your pizza just looked really good in the first place and then you get buried because you're not being, you know, progressive enough with I'd rather they spend more time washing their hands that it takes into account to show like, oh, it's a vibrant restaurant that's actively working on its menu. But yeah, I know I want better pizza. I don't care about the photos agreed to give me smells in the app. Every time I'm on, we talk about this. I need internet transmitted smells. Get on it. Back to fellow vision. Get on it. Every attempt has been a failure, but that doesn't mean you shouldn't keep trying. Keep working. Yeah. All right, folks, what would you like to hear us talk about on the show? We get great stories like this one from our subreddit. So go join in on the fun over there. You can submit stories. You can vote on them. You can talk to each other about the stories all at daily tech news show reddit.com. We often turn to nature when we want to replicate a phenomenon like flight, cell regeneration, coloring. There's recent research into the Ribbon Tail Stingray, specifically how it gets its unique blue coloring. And it's more sophisticated than you might think. Thankfully, we have Dr. Nikki with us to talk about and explain what the research is looking for here and how the results could be useful for us. Sure, thanks. So we're pulling this from a preprint from the City University of Hong Kong. And it comes to us this week talking about one of the rarest colors in nature, which is blue. So some of our listeners might already know this, but true blue as a pigment in nature is really uncommon. And if you see something in nature that is blue, it's actually not usually created by pigment, but instead created by a nano structure. So if you see a blue jay or a blue morpho butterfly, for example, they're actually brown covered by colored by by melanin, but they have a specific structural pattern that causes light to scatter and then appear blue in your eyes. So you get to see like this blue reflection similar to the effect of a prism. And in other animals, there's fish, for example, refish that are very brightly colored blue, and this is created by a similar effect created by cells containing organized, crystallized compounds that were also like scatter the light. But in this study, the blue coloration in sting rays has a newly discovered unique structure. OK, so if you've seen these rays before and, you know, obviously blue is in the name, they're called the blue spotted rib and tail ray. You can tell that they're blue. So what what have we found here that's new? Well, what's new in this study is that they're analyzing the nature of the sting rays coloration. And they actually found that in these really bright electric blue spots, there's not these like sacks of crystals like you would expect, which are called photonic crystals. And these are just kind of commonly found in nature. And the photonic crystals are the ones that they're only visible from a certain direction. So if you think about your your morpho butterfly, if you look at it from the side, it'll be brown and then you look at its face on and it'll be blue. Interesting. But to the surprise of the researchers looking at the sting rays, their blue coloration was not iridescent like the butterfly. And you could also see it from all different angles. OK, so all right. So so what what is unique here? What how how is this working? What's going on? What's happening? So I'm leading. Yeah, right. I'm like leading you guys onto it. So these researchers actually took a really a closer look with a scanning electron microscope. So they took a very, very, very close look right into their cellular structure. And they found these specialized cells within those blue dots that had a unique arrangement of nanospheres containing nanocrystals. And what's the most interesting about these structures and also what differentiates them from something like fish is that they're arranged in a nonuniform manner. So usually you'd expect this to result in a total scattering of light all over the place and you wouldn't have a lot to see. But actually this unique structure creates this blue color through random light scattering and it's still visible, but it loses the iridescence because it's not uniform like in these other animals I mentioned. So imagine like a jar full of marbles. That's kind of what these cells are holding and creating pigment or not, sorry, creating color from not pigment. And this is called a photonic glass ultra structure. And this is actually the first time that this type of structural coloration has been described in nature at all. We have these, but they're manmade. And it's also the first time that this structural color has been described in elasmobranchs, which is rays and sharks. So that's really exciting to me, at least. I I this is mind blowing to say something is blue and and Dr. Nikki's like, it's not blue. You just think it's blue because it's created blue because of the way, you know, animals, light and color works. So so knowing this scientific research, you know, and and and the sort of breakthrough here, what would be future applications with this knowledge that that we could use? Yeah. So the cool thing about this is because it's not a dye and it's a structure, it's a really interesting avenue for exploring the manufacturing of colors, because it won't fade. It it's just a structure, so it will just continuously reflect light. And you don't need additional chemicals to produce it other than whatever it takes to create that structure. So that's really interesting for manufacturing and all the photonic glass structures that we know of right now are manmade. So this new biological perspective gives us some more flexibility and potentially would inspire different ways to produce color and potentially on flexible materials like skin. And the difference between this and what's what a butterfly does is it's blue, whatever direction you look at it. Is that right? And it's not iridescent and the cellular structure. So in the butterfly, it's literally the physical structure of those scales that they have on their wings. And this is a specific cell type that like creates these little beads, if you will, microbeads and micro crystals in this random arrangement that reflects the light. So it's a little bit of a different biological process, but with a similar result. I think it's a fairly well known at this point and it may even be a myth, to be honest, I hear it so much. But, you know, supposedly royal blue comes from the fact that it was so difficult to create blue. No, it's true. And that indigo itself as a dye was like incredibly valuable, like as valuable as gold in the day. So, you know, we we've gotten better at blue. Blue is not so so so hard to make anymore. I'm not sure how much how much advance in technology of making blue we need. Oh, you can always use the technological advance, right? You know, indigo is not, you know, the sky being blue, you know, I actually went to a university in Toulouse where they're very famous for the blue de lectur, which is this one of these kind of stories about it. It took so long to figure out how to make blue and it comes from a yellow flower and it took a long time of trial and error to try to figure out how to chemically alter the dye to make it turn blue. But you'd be surprised about all the applications for blue. I mean, there's a lot of applications in lasers, actually, not specifically, but the specific glass structure and technological stuff and things like fiber optics with this kind of glass structure. So there's a lot to be done. Now you're telling me I might get faster internet because of this. Stingray internet coming straight to your eyeballs. Stingray is a great name for an ISP, right? Oh, there we go. Copyrighted Tom right now. Yeah, I feel like there was a stingray. What was that that like? No, it was ricochet. That's what I'm thinking of. This is so cool. Thank you for for bringing this story to us. That is really interesting, Dr. Nicky. Yeah. And I'll thank the researchers who send it into me and who requested I bring it up. So if you're researching something and you want me to bring it up, just send it to me. Yeah. All right. Let's check out the world. This is a real nice one that comes in from Chris, who says, I've been a listener for many years. Finally back DTNS starting in 2020. Thank you, Chris. I looked back recently and realized how much value I was getting from the show. I figured I'd give a little back while I can. I've been programming computers for more than 40 years and been in the tech industry my entire professional life. I've witnessed massive growth, rapid change and both the thrill and the chaos that comes from both. Staying on top of technology today is difficult. New products being introduced faster than ever before. The impact of some products is increasing more than we've ever seen. I often find myself busy with demands from both work and family eating up my time. So staying abreast of the news has become increasingly challenging. DTNS is my lifeline, enabling me to do to better understand what's going on in the world of tech without having to sacrifice my other responsibilities on top of that. Tom, Sarah and the team accompany me on my occasional commutes to the office on those days. Your voices keep me sane and otherwise insane Boston traffic. Can't thank you enough for that. Thank you for everything you do. And here's to many more years of DTNS. Oh, thank you, Chris. That's so sweet. That is so heartwarming. Yeah, thanks, Chris. It is exactly what we're trying to do. We are like we will spend the hours sorting through all these stories and trying to understand them so that you feel more informed and don't have to spend hours doing it. So that's that's exactly what we love to hear. Really appreciate you taking the time to send that, Chris, because it's easy for all of us to send an email when we're upset about something or we want to fix something. But taking the time when you just want to say thanks for something is very much appreciated. So thank you indeed. Yeah, heartwarming indeed. I feel very indigo right now because of you, Chris. I'm not blue. I'm happy I'm a special kind of blue from all angles. Sky blue. Speaking of sky blue, Dr. Nicker and Nicky Ackermans, you always make us feel sky blue when you're on the show. Thanks for being here today. Let folks know where they can keep up with the rest of your work. Absolutely. Everyone can find everything to contact me on my website, nickolackermans.com. If you want to send me a DTNS article that you want me to cover, I might consider it. If you send it to my email that's linked over there. I'm also on Twitter at AckermansNichol and on bluesky at nickolackermans.bluesky.social. And indeed, Charlie and Columbus and Roger both reminding me that Stingray is the name for the fake cell tower. That's probably why it was in my head as an ISP. So since a fake cell tower that's meant to capture technology is called a Stingray, an ISP probably isn't going to want to call itself a Stingray. Yeah, thank you for the reminder, both of you. Patrons stick around for the extended show. Good Day Internet. It's not just food menus. AI images are sneaking past peer reviewers in scientific papers in ways you think would be easy to spot. Like really easy anatomy of a mouse, my friends. Like, how did you miss that? But we'll explain. We'll explain what's going on with that. Stick around. You can catch our show live Monday through Friday at 4 p.m. Eastern 2100 UTC. Find out more at DailyTechNewShow.com slash live. We're back tomorrow talking about the Video Game History Foundation's unique approach to recording video game history with none other than Scott Johnson. Talk to you then. The DTNS Family of Podcasts. Helping each other understand. Diamond Club, I hope you have enjoyed this program.