 This 10th year of Daily Tech News show is made possible by its listeners. Thanks to all of you, including Dustin Campbell, Tim Deputy, Brandon Brooks, and our lifetime supporter Casey Heward. Thanks, Casey! On this episode of DTNS, can AI successfully apply for a lot of jobs that you want at once? Well, someone actually has tried this out. More details are emerging about Cruz's autonomous missteps, and Dr. Nikki Ackerman tells us all about the latest research in coastal climate change. This is the Daily Tech News for Tuesday, November 7, 2023. From Studio Secret Bunker, I'm Sarah Lane. From Columbus, Ohio, I'm Rob Dunwood. And from Alabama, I'm Dr. Nikki Ackerman. And I'm the show's producer, Roger J. Well, the first day of the Epic Games vs. Google antitrust trial included both sides giving opening statements. There were a couple witness testimonies as well. Epic's lead attorney Gary Bornstein claimed that the Google Play Store accounted for 90% of app installs back in 2020, that was when the lawsuit was filed. Google's attorney, Glenn Pomerance, argued that Google is an anti-competitive, noting that Google and Apple compete against each other. We are going to talk a lot about this with Scott Johnson tomorrow because he is following this case. So for now, on to the quick hits. X's API access requirements, X being former Twitter, have changed as of late as of many things on the platform causing some researchers to cancel, suspend, or even change more than 100 studies about X since the feature allowing free academic access to 10 million tweets per month from the API doesn't exist anymore. Researchers told Reuters that current restrictions on data gathering have hampered fact-checking information during real-time events such as the current war in Israel. Researchers also say they fear being sued by acts over findings following the Center for Countering at Digital Hate being sued by the company after it published critical reports about X's content moderation back in July. Samsung's S-Pen Creator Edition stylus for tablets that the company announced along with the Galaxy Tab S9 series back in July is now available to buy in the US for $99.99. The S-Pen Creator Edition is compatible with the Samsung Galaxy S21 Ultra and higher, the Note 10 and higher, the Galaxy Tab S series and any PC that already supports the existing S-Pen but no Galaxy Z or Z series of folding devices. Last week Microsoft announced Xbox Game Pass Ultimate benefits would be stripped from most of Microsoft's 238,000 employees starting in January and boy were people not thrilled about that. Now the company is backpedaling saying we're reversing this decision with Phil Spencer, Xbox Chief, saying after looking into this more of the team just want to confirm that no change will be made to Game Pass availability in 2024. If you have access to the Game Pass offer today you will continue to have access, I appreciate the time to get up to speed and sorry for the questions and confusion created. And thanks for supporting Xbox. Bloomberg's Mark Gurman reports that Apple is pausing development on iOS, iPad OS and Mac OS offer to iron out glitches in the code. Microsoft said the delay was announced internally to employees last week after a number of bugs were discovered in early versions. This is unusual but not exactly new. Back in 2018 software engineering chief Craig Federighi pushed back a handful of unreleased iPhone features until the following year over bug concerns. Google first announced its magic editor tool for its Photos app at Google I-O 2023 earlier this year and code in the latest version includes specific error messages if you ask magic editor to do certain things. For example, edit photos of a driver's license or ID card, a receipt, an image with personally identifiable information, human faces and body parts. Magic editor could also be getting text prompt based generation abilities as such could also refuse to generate images based on some words mentioned in that prompt. And those are the quick hits. Rob, let's talk about what is happening at GM's Cruise. So GM's autonomous vehicle aka RoboTaxi Division Cruise recently had his California operating driverless permits pulled over safety concerns and subsequently halted all driverless autonomous operations where it does business. Now other issues with the company are being reported such as problems with the number of times human operators must remotely take control of vehicles seemingly every four to five miles of operation. Cruise also seems to have difficulty recognizing children around its vehicles. Yeah, so if you're like, what? Tell us more about the kid thing. So Cruise said in a statement to the Intercept that its vehicles sometimes temporarily lost track of children by the side of the road during simulation testing. It also said that the problem has since been fixed, was only seen in testing, wasn't an issue on public streets, although that's still going to give a lot of people pause because we're talking about kids, right? The California DMV has given Cruise the steps needed to reinstate suspended permits. So they could be reinstated, but they have not yet. Nikki, what do you think about this? I mean, obviously autonomous vehicles have to do quite a bit of testing and some companies are farther along than others, but you know, you hear stuff like this and of course people are going to be worried, right? Yeah, I think I'm thinking the same thing as everyone else is that that's a little bit concerning that nobody thought about that before, you know, of course it happened only in testing, but that means it could happen in all of the other cars that were in service before that and I would be a little bit concerned. I'd want to check that code and check it for not only children, but something we mentioned a bit earlier, maybe people in wheelchairs or dogs or all the other things you would not want to hit in an autonomous car. That was my first thought too is, oh, well, if we're talking about children, maybe it's a height thing. It also might be a darting around in a way that kids do, that adults often do less. You know, I don't have all the data there, but yeah, you can't be like, oh, sometimes our cars, you were like moving weird. Well, and at first, when I when I first saw the story, I thought, well, that's an issue because anything involving a minor is an issue, but this is not within the vehicle itself. This is the car being able to, you know, identify a kid on the side of a crosswalk, for example, that kind of stuff. Now, again, company says this is just testing, but Rob, I don't know, you know, where do we go from here? Children, small children hyped up on sugar are the most erratic beings on the planet. This is where you really need to make sure you've gotten stuff buttoned down. So just, you know, just some additional background on this. And one simulation, the company couldn't rule out a scenario where the vehicle strikes a child and another specific test drive, a vehicle detected the child, it was a child test dummy, but it still struck it with the mirror 28 miles per hour. Now, I don't know, most children are pretty small. So the mirror is probably going to be at head, height, shoulder, neck height, 28 miles an hour. You know, for me, that's just going to fold it in and I might have a bruise for a child getting hit in the head like that. That's pretty bad. So I absolutely hope they are, they're doing everything they can do to get this stuff corrected before they get these cars back out on the road. So much of this I also feel like is, I mean, a lot of it is, okay, so public hears about this, public freaks out. I mean, we're not freaking out on the show right now, but we're talking about some very serious concerns. There are so many concerns with driverless tech in general from, you know, how companies are changing the way that cities work and are they safe? And, you know, there's always the, the kind of argument of like, well, human drivers make mistakes all the time and driverless technology, if working appropriately, should cut down on a lot of that stuff. The human error, we can, we can take ourselves out of the equation. But again, you know, humans are making the code that are, that are making these, these vehicles do their thing. So I think it makes a lot of sense for crews to say, you know what, let's just pause this whole thing. Let's go back to the drawing board, fix some issues, be transparent, I guess. I mean, assuming we know all there is to know about what still needs to happen. And that makes everybody safer in the end in a perfect world. I would trust crews more if they would have done this pause before they got shut down in California. The fact that, oh, we're going to shut down and build trust back because something else happened and we're also have these issues that are going to be fixed is like, OK, well, they, they probably would have kept running if no one would have ever shut them down. So I just, I hope that, you know, like the, the reads on this, you know, people are writing the words, but the way these things read, it almost sounds like it's kind of nonchalant. Oh, it's just a security issue and testing. We'll get this right. And that kind of sounds to me like, you know, once I figure out how to flat my arms fast enough, I'll be able to, you know, take flight. No, you need to actually create the software, test the software and make sure that it doesn't hit little people, you know, children. You do not want your cars doing that. Yeah. How did this get through like the safety protocol to release the car in the first place? Like, how are you not checking? Obviously, you can't predict problems that you don't have, but like, surely you should test your cars on children before you let them out into the wild, both the children and the cars. Yeah, you know, I was I was trying to gauge the temperature of how people felt about crews this morning, just kind of looking into the story. And there's a lot of knee-jerk reactions of like, you see, we'd never should have done this. It's out of control. You're going to hurt people. And, you know, to be fair, you know, crews are saying, this is a child was not struck. We tested, you know, dummies that were of child-like height and, you know, qualities and some stuff happened and we're working on it. But yes, Rob, you're right. This may never have come to light, even if the company cares very much about safety first, you know, because, you know, why, why wouldn't you with a technology like this? But it probably would have been something that was quietly handled behind the scenes. Indefinitely, maybe, but at least an attempt to handle it, except that a lot of stuff is now just a public concern. So can't get enough Android in your life. Then listen to Android Faithful. Every week, Android Affectionados, Ron Richards and Wente Dow bring you the latest Android news and information. You can watch it live Tuesdays at 8 p.m. Eastern, 5 p.m. Pacific and subscribe to the feed at Android Faithful. Dot com. Well, documenting the impact of climate change on coastal environments along the Atlantic Ocean is not easy work. You've got satellites, aerial surveys offer some data, but only so much. So when you need more precise information, you need to put a sensor at the water's edge in the intertitled zone. But technology is stepping into at least help bring that work forward a bit. Nikki, you've been you've been looking into this. So explain what's going on with this research, the sensors that they're using, you know, walk us through this. Yeah, absolutely. And I'm excited about this. We don't usually do ecology related stories. So I'm happy to bring this to DTNS. So I was looking through this and I noticed this great intertitle story. So researchers from the University of Porto are currently working on this project. They want to deploy up to 2000 sensors all along the shores of the Atlantic coast, both in Europe and in the US and South America, all these places at about 320 sites. And their goal is to understand how climate change is impacting the Atlantic Ocean. So like you said, Sarah, we've got satellites for data and aerial surveys and buoys in the water, which is telling us, you know, temperature changes and things like that. But the intertitle zone, so the zone between the coast and the shore and the ocean is kind of in between those areas. And it's kind of a little micro ecosystem that we don't have a lot of data on. And we need to understand how climate change is impacting that area as well. And so the Atlantic Ocean coupled coastal temperature and biodiversity observation network has a very long name and is in charge of this project. And they're trying to protect marine biodiversity while the ocean is changing. So how are the researchers actually addressing this problem? Well, they started in 2008 with a project that is very DTNS. They made some robo limpets. So limpets are a little like cone shaped mollusks that live on the rocks. And they took all the electronics they needed for this sensor and stuck them inside a limpet shell and glued it to a rock, which I just think is wonderful. So they're about three centimeters wide, a little bit bigger than a water bottle cap, and you've got a temperature probe that takes hourly measurements and then you can download data onto a smartphone. The problem with the robo limpets is that some of them got washed away by the surf. This was in Portugal where you have really, really strong waves and tides. And also the locals eat limpets, so people were probably picking them up and harvesting them. And we were like, oh, man, just a rock. Can you imagine you find a limpet to go eating? It's got like tech in it. And you're like, it's robots. They're taking it over. It's amazing. So OK, they have a new and improved version of this now, which is what this story is focused on. And these ones are they actually just drill a hole in the rock and place this sensor in there and then cover it over with epoxy. So they don't have a limpet shell anymore, which is a shame, but they are way more practical and way more hidden. And they have enough data storage and battery life to last for about a decade, which is pretty impressive. And oh, go ahead. No, I'd be I'm agreeing with you. That's that way longer than I thought. I thought you were going to say like for like 10 days. No, yeah, 10 years of data. Can you imagine? And they've put this on every single continent. They've got some in Antarctica. And if you wanted to participate in community science, you could buy your own for 50 bucks from their startup, Electric Blue, which is a great name. And you could also contribute to the project. So Dr. Nicky, what are these sensors? What are they actually finding? You know, what information are we getting back from them? Yes, wonderful question. So this huge sensor network actually has revealed that the intertidal temperatures vary so much, they vary really even if you're one meter away in one area that's in the sun and one area that's in the shade at low tide, high tide, there are all these different variations happening. And if you're at low tide and you're exposed to the sun, you get a really high amount of heat as opposed to if you were in the shade as a mollusk or whatever you are in the intertidal zone. And until now, we kind of assumed that this meant that if you went further towards the poles and the areas were getting a bit colder, that the intertidal zones were also getting colder. But it's not necessarily true. And so that was proven by these sensors that like you're still very hot if you're in the sun, even if you're up north. This is actually a really important finding because those cooler areas might serve as a refuge for organisms that are really sensitive to climate change creating rising temperatures. And another really important finding from this network of sensors is that we thought that the most vulnerable zones for these rising temperatures would be the tropics. That would be kind of the logical assumption. But it turns out they're a little bit further out from the tropics. So now we can focus more of our conservation on those areas and kind of make a more sensible decision about how to preserve these zones. I mean, wouldn't the zones also I mean, it would the temperature would would depend so much on how shallow the water might be. I mean, I think a lot of us have been on vacation where it's like, wow, you can walk out for a while before it even gets deep in the water. You know, where, you know, other other zones, especially the Pacific Ocean, which is near me, it drops off really quick. I mean, you don't really have any shallow water at all. Yeah, that's why it's so hard to measure. There's so much variation in these areas, especially with the tides that that's why we weren't getting any of this information before with buoys and with, you know, like just drones or whatever other kind of sensors you could have there. So any future plans do we have? Yeah, and I'm really excited about their future plans because they're planning to store all the data from all of these two thousand loggers on an open access platform so that they can improve access to scientists in lower income countries and kind of just expand everyone's knowledge about how climate change affects these really fragile ecosystems. And I love that. That's like such a wonderful bow on this story. That's really cool. Increasing access. I love it. Yeah, excellent. If people want to find out more about it, where should they go? Well, you can visit the website, electricblue.eu. That's for if you want to get one of these loggers. They have some really cool ones that are shaped like muscles, which I just think are cool. And I think from there, you can find about all about how this project is going and eventually when they plan to release the data and then maybe you can access it yourself. Well, we mentioned, thank you, Nikki. That's awesome stuff. We're going to be talking to Nikki a little bit more about science in GDI, by the way. So do stick around if you are a patron and talk about genetics and we'll be taking some questions live. But for now, we mentioned something about perhaps using AI to help you get a job. We know AI is kind of used for everything now. It's such the buzzword. But if you are a software engineer named Julian Joseph, you use it a little bit differently. So I'll back up a little bit. If you are applying for a job and you have done that recently, it can be a numbers game. Wired has an article covering the role that AI bots can have in helping people apply for jobs. People are already doing this, writing job descriptions, finding candidates on the other side. Both both are definitely in use. But Joseph, who used an AI bot called lazy apply after being laid off, had some interesting results. So lazy apply is sort of what it sounds like it's an online job application service that uses AI to automatically apply for jobs, apply to jobs that you are interested in with a single click. So you pay a price, in this case, $250. And Julian Joseph got a lifetime unlimited plan. Hopefully he will not be searching for a job for his entire lifetime. But you get the idea, scored him 20 interviews after applying to 5,000 jobs. Now, 5,000 jobs, that's insane. But hey, 20 interviews, that's also a lot of work, again, for both parties. So you figure for the price, you certainly got at least a foot in the door. Article highlights that the job's application services are as a growth industry, with some using AI, others using a mix of both human and AI to fill out applications for clients. Recruiters, though, had mixed reactions. Some felt using AI meant that the candidate was very serious about the job, using all the tools afforded to them. Others said, yeah, I don't know how that resume even get to me. The person didn't even do anything human ask. Hopefully that article didn't make him lose his new job. I haven't applied for a job recently, but I have been applying for a new apartment. And, boy, would I like AI to help me out with that. That would be a great application for that. Because a lot of it is like, yeah, you do have to, we don't have to, but to customize it to kind of, you know, we all sort of know how that works. It's like, oh, are you having a conversation with a person about your pet? You want to make sure that they think that you having a pet is not going to be an issue and you sort of craft the application a certain way. I can see this working. I mean, I haven't been in the job market for a while where I was filling out applications. Dude, it sucks. But I would I would have to imagine that, you know, if you only got 20 interviews from 5,000 applications, how closely matched to any of those jobs was he? Because that that is a ridiculously whole ratio. I'm not surprised. So the article makes I didn't I didn't write it in. But when he filled out applications himself manually, he he would only have to do around, you know, 200 or so applications to score around 20 interviews. But because it is AI powered, he didn't have to do anything like this. This service applied, applied for a thousand jobs overnight while he went to bed. And so it's the trade off. I mean, there's a trade off. You're in fact, the article, the article mentions that in many ways, it's just brute force application. Throw as many applications out into the into the job market as you can and in a hope to score something, you know, and it's I only know this because my wife is is is currently searching and I actually have a couple of friends who are as well. And the process is very grueling. It is not what I would have expected having applied, you know, look having the last looking for work around 2013, 2014 and it is very much a numbers game. What you were saying initially in the article in the read as Sarah, you really need to just crank these out because there are so many positions in so many of these jobs, they're the companies, they use machine learning to sort the initial applications. Right. So they look for key words. They look for time frames. They look for things that that they want and that helps reduce the number of physical applications a human person has to go through. But I mean, machine learning AI is now part of the hiring process, whether you like it or not for me, for me, either side. Yeah, as someone who's hiring right now and also who like started my job like last year, I pretty fresh in it. It's a little bit different for academia because we need to provide so many documents. People ask for personalized stuff. That's the new fashion now personalized statement of research personalized statement of this. So like you can get chat GBT to write it, but you still have to like adapt it. I wouldn't encourage getting to write it, but people do it. The problem is I'm like on the fence because on one hand like, well, if it gets your application to my eyes, maybe that's better. On the other hand, I really don't want to read your statement that was written by chat to be that shows me that like you don't care. But otherwise maybe I would have never seen your resume. I don't know. It's kind of a, I think right now a lot of people are be like, that's unethical, but it depends on the job. Totally. Yes, definitely depends on the job. I mean, what you're describing Nikki is exactly. It's like, hey, everybody's qualified here, but who's the best fit, you know, let's find out more about you as a person. I'm going to guess that something called lazy apply is not going to be the best way to convince somebody that you as a person, you know, are, are, are, are it, and you're right, Rob, of all of those applications to get 20 interviews. I mean, it's still, that sounds very time consuming to me 20 interviews. Maybe they're not the, like the best 20 interviews top five. Right. Yeah. It's like, I mean, sometimes it's like all you really need is the one, right? You know, maybe there's just the one job that you want so much, but again, yeah, it depends on, depends on the kind of work and the career. On someone's desk, then sure, why not? After that, then you can personalize it. I mean, I used to argue with friends of mine who would, these are the days of like printed out resumes that you would send or like fax somewhere. But a friend of mine, he swore up and down that having like a headshot on his resume helped him get more interviews because people Oh, that's illegal. couldn't not think of him as a human, you know, because now they've looked at this person and so they cared about him more than if he was just a bunch of words on a page. And I was like, was he a white dude though? Yeah, it was just, well, yeah. I was about to say he was, he still is, he's around. He's pretty successful. So anyway, yes. Not a one stop shop going on here, but Kudos to software engineer Julian Joseph using Lizzie Apply and getting some interviews and we hope you got the job you wanted. All right, let's check out the mailbag. Back on last Friday's show, Nate Langston joined us and mentioned YouTube not doing picture and picture on iPhone and Nate says, it's really annoying. Matthew wrote in and said, it actually does. When the desktop site is loaded, not only does picture and picture work, but you can also listen to music with the screen off this way by restarting playback from control center after turning the screen off. I feel like I saw, I've seen, I've seen this, someone talk about this tweak before. Didn't, it was not always the case, I believe. It's something that YouTube now does, but didn't in iOS before. Not sure, I'm an Android guy so I've always been able to do it. But we had another mailbag item as well. Freedlay in Australia emailed about last Friday's topic as well on YouTube Premium. They write in response to Nate's opinion on YouTube Premium, while I agree that it is quite expensive, especially for single accounts, I didn't hear mention of the cost advantage for the family plan. As a family of four, considering the usage of two teenage boys, YouTube Premium has become one of those expenses that is hard to justify giving up. Yeah, I don't pay for YouTube Premium. I know people who do, and often they are on family plans and say, how can you live this way, Sarah? I don't know, it just depends on your YouTube usage. What about you, Nikki, do you pay for YouTube Premium? I don't, I'm like the last person to pay for something. I'm still on my parent's Netflix account, don't tell Netflix. Nobody's on Netflix. Yeah, it's fine, you know, so it goes, so it goes. All right, well, as promised, Dr. Nikki, as soon as GDI starts, we're gonna talk to you a little bit more about some science questions. So everybody who's a patron is gonna listen to GDI, get your science questions ready. Dr. Nikki, she can't answer everything, but she'll do her best. I'm gonna try, I'm gonna try. Yeah, yeah, it'll be fun. But in the meantime, let folks know what they can keep up the rest of your work. I will do that, yes. So on GDI, we're gonna talk about genetics, and I'm using that as a segue to say that I am hiring a postdoc in my lab in Genomics. If you do that, you can find more information on my website, which is nicoleackermans.com, and you can find all the other information about everything I do there. I'm also Ackerman's Nicole on X, and Nicole Ackerman's on Blue Sky. I said it right today, and I'd love to chat with everyone on all those platforms, cause I love you guys. Yeah, so Nikki's gonna be answering some science questions from the audience, and you know, if you're a patron, do stick around, it's gonna be fun. Good day internet. Comes up in just a couple of shakes of a lamb's tail. Give me a challenge, I'm ready. So you can also catch the show live Monday through Friday at 4 p.m. Easter in 2100 UTC. Find out more about dailytechnewshow.com for size live, we'll be back tomorrow with Scott Johnson. Talk to y'all then. This show is part of the Frog Pants Network. Get more at frogpants.com. I hope you have enjoyed this program. Hehehehe.