 Daily Tech News show is made possible by you listening right now. Thank you, Tim Ashman, Johnny Hernandez, High Tech Oki, and Ellen Stearns. On this episode of DTNS, Canadian school boards sue all the social networks. Cops can shoot GPS trackers at fleeing cars now. And does China have an edge in AI talent? How do we measure it? This is the Daily Tech News for Friday, March 29th, 2024 in Los Angeles. I'm Tom From Studio Redwood, adjacent, I'm Sarah Lane. I'm the show's producer, Roger Chang. And I'm Andrea Jones Roy in New York City. Hello. Data science professor Andrea Jones Roy, thank you so much for joining us today. Thanks for having us. And thank you as I coughed up a piece of Apple for jumping in and introducing yourself. You're a total pro. I didn't leap to, you know, save you from the Apple, but I did leap to say my own name. So we'll go with selfishness. Well, while I try to cough that up, let's start with the quick hits. The US succeeded in almost killing network equipment and handphone maker Huawei when it imposed restrictions on supplying it with parts or intellectual property from US companies. Now, Huawei has found a new path, riding on some creative thinking and national patriotism with the company's handsets, rising in sales in China while all other brands foreign and domestic are generally falling. While we also just announced its fastest growth in four years with revenue up 9.63% mostly coming from the consumer business. The New York Times sources say Metta will add some of its multimodal generative models to the Ray band smart glasses starting next month. Features have been available in early access. Since December, they can do things like translate, identify objects like pets or art or landmarks. The glasses support English, French, German, Italian and Spanish. If you happen to be a US Google podcast user, you have a few more days to transition to a new podcast service before the Google one shuts down on April 2. Google recommends its users export their subscriptions to YouTube music. Of course they do, but you can also export your feeds as an OPML file and then add them to any other podcast app as well. Users outside the US apparently have a little bit more time to figure out what they're going to do with a YouTube music support page saying Google podcasts will not shut down entirely until July. Hey, so I guess you could VPN too. I don't know how that works. One of the biggest growth areas for AI is companies making bespoke models that they can roll out to their customers and maybe a customer support chat bot or or a sales bot or some other function. But when companies do that, they take a lot of risk because they are now responsible for what that chat bot says. Microsoft Chief Product Officer of Responsible AI. That's a title at Microsoft. Sarah Byrd told the Verge that Azure customers who can't afford their own red teams to test AI instances can use some new tools to keep them safe to use for their customers. There are three tools coming out right away prompts shield, which blocks the malicious prompts that a customer really an attacker might use to try to fool the LLM into revealing information it's not supposed to grounded nest detection. I don't think that's a word, but okay, grounded nest detection, which blocks hallucinations, aka incorrect facts and safety evaluations, which looks for model vulnerabilities. Two more tools for Azure AI users are coming soon. One will be used for nudging models towards safe outputs and another to flag potentially problematic users. The safety features are immediately attached. If you're using a GPT for or meta llama to model, and you can point the other smaller models, you can point them at the other smaller models, if you want, as long as Azure supports that model. Hadnuki.com reports that Samsung has faced poor yields for 11 inch OLED displays that are meant for new iPads. Apple has reportedly had to transfer some of the production of those new iPads to LG display. New iPads are now expected to be announced in May. That's according to Bloomberg's Mark Irman. All right, big thanks to taxi cab over on our subreddit daily technology show dot reddit.com who alerted us to the story earlier this week. Four school boards in Ontario Canada are suing meta snap and bite dance for a combined $4.5 billion alleging that Facebook, Instagram, TikTok, and Snapchat create products designed to negatively affect children. The school boards alleged in a press release that students are experiencing an attention, learning and mental health crisis because of quote prolific and compulsive use of social media products. Snapchat and TikTok issued statements denying the allegations and arguing their apps are designed to combat such negative effects. Ontario's premier Doug Ford said he disagrees with the school board's lawsuits. He says look phones were banned in Ontario classrooms in 2019. So if they're using these things in the class to begin with, that that's a problem. We should already crack down on that. Andrea, I know you didn't teach high school, but you know, you've taught some some some folks fresh out of high school and in some cases. What's your take as an educator on this? Phones in classrooms are nightmare and I feel like that makes me sound like I'm completely out of touch and anti tech, but I promise you that the the students in class who are just on their phones and just zoned out scrolling are not doing very well in class. And so if you know just putting it on do not disturb and putting it away does wonders. I was one of those professors. I have opinions about this who didn't allow laptops in my classroom for a long time. But I do teach programming. And so that did actually become a little bit difficult to enforce. And so I caved on that, but it really is distracting. And if if I'm sitting in an audience and I have my phone out or my computer out, I'm not paying attention. I also last thing on this notice that something that I find very chilling for my own personal reasons is I would gather outside the classroom. I was at NYU like, you know, in these big classrooms, all the students are milling around waiting to go into our room. They're all on their phones. When I was their age, I was talking to my classmates asking questions, doing whatever, getting to know one another. And everyone is just sitting there silently. So it's not just in the classroom. I really feel like it's had a dampening effect on the culture in the classroom outside of it. But theoretically, that's what they're doing on Snapchat and TikTok, right, is talking to their classmates just over the social media. They might be shit taught. Pardon me. They might be talking, saying bad things about me during class. They all have that groups and we chat groups and all the rest of that's no doubt about that. But I mean, they're scrolling social media. They're not, you know, checking on homework answers, as far as I can tell. I mean, it sounds like, and it's interesting that Ontario Premier Forbes said he disagrees with the school board's lawsuits going after this, I guess, in the manner that they did. But, you know, I don't have children. I was a child once and there was no social media at the time. So, you know, I can't begin to understand how much of this is either a distraction or, you know, just a downright harm on certain kinds of kids and their learning abilities, right? Like the idea is to make kids be, you know, as safe as possible while they're learning, all is good. And as a teacher, you have a really hard job anyway. But, but, you know, banning phones, laptops, you know, basically anything that can be online outside of school. I mean, that's one step. But if a kid is having a hard time focusing, learning, understanding things because of this, and that can be proven, that I think this lawsuit is actually pretty interesting. Yeah, and I think it's worth refocusing this. The ban already existed. That's not what's at issue. What's at issue is that the school boards are saying that these companies design their products, possibly in a negligent way, maybe not on purpose, although the lawsuit is pretty strongly worded, in a way that damages children. And it affects them in the school. And that gives the school board the standing to sue. Roger, you are a parent. Your kids are a little on the young end of this, but they'll be heading towards this situation soon. How do you feel? I, one, I think your phone should be should be blocked on school campus, campus grounds. But I think social media has created a kind of a parallel social network universe that that that happens in a way that I think will negatively impacts can negatively impact children because they're there. It's a way of them communicating with people they feel comfortable with. And one of the things you learn in school and someone who's bullied a lot in school, you kind of have to learn how to manage relationships that aren't necessarily always in your favor, right? And it's an unfortunately, you know, one of my children has ADHD. And so, you know, using her tablet, although not social media, because I've put the kibosh on that for now. But she does talk to one of her close friends over, over through Roblox. My youngest just yesterday wanted to do the same thing. We said no. And she threw a tantrum for 15 and like the legs and arms flailing in the screaming. And I'm curious to know how this turns out because we do block certain things, for example, alcohol, cigarette advertising from a certain distance away from where from children schools are. This could be seen in that way, although because it is a digital, you know, a digital platform, you don't have a, there's no physicality that you can, you can put up. But I mean, I don't know. You're saying it's possible that that Facebook and Instagram should be regulated like cigarettes and drinks and alcohol? I mean, it's it could very well be if not, if not, if not in Canada, somewhere else, I definitely want to know what the evidence of the harm is, right? I feel like there's a lot of like, we think it's bad. And so let's ban it. And I'm not saying it isn't bad, but I don't know that the due diligence has been done and the studies have been done to show how the harm is perpetrated. Because you may introduce, it's possible you introduce more harm by banning students from social networks, which cause them to use them out in ways that aren't detected and pretend to be adults and things like that, that would be worse. Then if you're like, oh, you know, the harm comes from this, let's focus on stopping that kind of behavior from head. There's a new book out by a scholar at NYU named Jonathan Haidt, and it is specifically about Gen Z and smartphones. It's not social media, but it's about smartphones and basically summarizing the research that he chose to focus on that says, yeah, it's pretty bad for learning. But Tom, I'm with you that it's, you know, there's a lot of gradients between ban versus no ban. One is how old are the students? You know, I could imagine something like preschool first grade being in person and interacting matters more than in my class, where if you're an 18 to 22 year old, part of learning to be an adult in society is managing, I got to pay attention to this and I have a phone in my pocket that, of course, if an emergency happens, I can attend to, but I also am learning to regulate my own attention. And so I think in my classroom, it's very different from little kids. But there's also the class itself, right? So a couple of people in the chat have mentioned, look, you know, what if you, as the teacher, train students how to Google more effectively? And I would do this, I'd say, hey, we're coding, you got an error in your code. Here's how to learn how to figure out what the error is. Use the internet, use chat, GBT, and we do it live. Whereas if you're reading poetry and reflecting on it, then maybe you don't need it in front of you. And so I think there's more to it than just a black, white band, but there is research that mostly is is suggesting from what I have seen that it's mostly not great for children. Well, and we're talking about 13 plus, I'm guessing here. Like, I don't I don't think these social networks allow, I know there was, yeah, I know there was some talk of having Instagram provide kid friendly versions of accounts. But but we're mostly talking about high schoolers here. So I I do think it's it's not as as simple as like, you design these to harm children. And so you owe four and a half billion dollars either. Like, I feel like that that doesn't necessarily fix the problem either. But I guess it's going to make a lot of parents and teachers feel better if they win this. I'm not sure if they'll win it. I'm not sure what the the I'm not well versed enough on the Canadian Canadian law here to say, I mean, it's sort of reminding me of the lawsuit against Jewel, right? That was saying you're marketing nicotine to children and that like the the ability to specifically, like you were saying, demonstrate that it was done out of malice or out of negligence and has specifically had this, you know, we can identify the mechanism for which it's doing something harmful to children, maybe. Otherwise, it does seem a bit misguided to go after it's just because something's bad for students in the classroom, maybe on average, not all cases doesn't necessarily mean that the provider of it is at fault, but that's the basis of all social media and internet regulation debates in the US, too. So yeah, answer. Well, the drive reports that old Westbury police in New York, Long Island, Hamlet, I suppose, is using vehicle mounted projectile launches to fire GPS trackers embedded in foam darts with sticky glue in them. Stay with me on the story because it is real. The darts are made by a company called Star Chase. The idea is to reduce the need for a dangerous high speed chase. If somebody happens to be, you know, going crazy down the road and a cop is following them while tracking the suspects, you would only need to chase them long enough to fire the sticky dart to, you know, get on the back of the car and then figure out where the car was going and maybe, you know, have a safer arena to to make sure that the car doesn't go any farther with minimal issues for anybody in the community, plus the driver, plus the police who are following them. Now, this is not the first time this has happened. If you haven't heard about this before, it has happened at certain police departments in Michigan, Tennessee, Texas and Washington state. So how do we feel about this idea? Now, you know, and let's get our jokes out about, you know, if you live in LA, you love a good car chase because you have a lot of them. But this will never fly in Los Angeles. It is our God given rights as citizens of Los Angeles to watch high speed car chases on the news. It is our major form of entertainment. I don't even live in LA. Yeah. Yeah, I mean, for my life. Yeah, totally, totally. That said, you know, when it really comes down to it, a lot of people's safety is at risk. When something like this happens, you know, absolutely about the helicopter. It's about all the people on the ground because things are happening in weird ways. You know, the you know, driver is acting erratically, et cetera, et cetera. So let's let's let's let's make it very clear. When there's a high speed chase, you have two cars that can run into folks and kill them. If you hit them with a GPS tracker, you have no high speed chase because there's no one chasing the suspect. You're just tracking them. So that's much safer for everybody. Yeah. Yeah. And I think, you know, it's case by case, but it's like that might be OK, we can get them around this bend because we know where they're going or we just wait till they get off the road and then, you know, go find the person, you know, who is the perpetrator at risk. I mean, every every case is different. This is, you know, it it raises a lot of questions. It raises questions about, you know, can you be trapped legally at any point? You know, because if you're if you're fleeing from a scene of something, you haven't necessarily committed a crime. You might have. You might not have. And, you know, that that that is one school of thinking. There's another school of thinking saying if you're doing something that is, you know, potentially harming anybody else, then, yeah, this is the right call. If you were to be tracked before or after an incident, that's not OK, you know, without, you know, a warrant or, you know, something similar. But yeah, it seems like this is pretty clear cut. There is a there is a Supreme Court case, U.S. versus Jones, not Jones Roy, but Jones that ruled that tracking a car without a warrant constitutes an illegal search. However, the ACLU has weighed in on this and said if someone is a valid suspect, right, which is usually what happens with a high speed car chase, the police saw them commit the crime or have have credible evidence that they committed the crime and they chase after them. In those cases where you could chase after them, it's fair to track them until you've apprehended them. And as long as you don't continue to track them after the apprehension or after the case, as long as you're not using it to continue to track them, the ACLU of all people is fine with this. All right, yeah. Andrea, what where do you follow on this? I mean, until the ACLU came along and was like, yeah, we're OK. I was like, this seems shady, right? I mean, my two thoughts go to where you were pointing out, which is, you know, how do you decide when to chase this person? How do you decide when to take this thing off this person? How do we know that we're not also going to secretly use these darts to track the person I'm stalking or someone else that I want to, you know, conduct a heist on later on? It just seems like it's ripe for misuse. But then my other question is, don't the now that the story is out and now that this is a thing that people are using don't people who've just committed crimes now know to look to see if their car has been tagged with a tiny dart or is that not as easy to detect as it seems? Like on the picture, it seems like figured out, right? You just pull it off. It's a little nerf dart. Yeah, you wait until you've gotten far enough away that the cops aren't chasing you anymore and then you just put it on the car. Yeah, I don't know. Maybe the glue, the glue, it's a heat activated glue. So maybe that makes it difficult to pull off. And if you stop long enough, then they come and apprehend you because it's not a high speed chase anymore. I also think, you know, if for some reason, you know, law enforcement was like, I want to track Sarah, you know, I think she might be about to do something weird. Then I'm not vigilant about this. You know, you know, I'm not saying that somebody's going to track me unlawfully. But if they were to and I wasn't, you know, in a high speed chase scenario, I might not notice. And that would be a legal issue. And that's not the situation we're talking about. It isn't right now. That's what the ACLU said is like, look, if somebody flees the police, they are by definition breaking the law at that point. And so it's fine to track them. Yeah. But if you were using this in the situation like you're talking about, Sarah, then it would not be OK. So yeah, it feels like, Andrea, you're right that you would know if you're like, oh, I'm fleeing. I bet that and I heard the bonk. You know, I don't know how. They're made of foam. So maybe you can't hear it. Or you look at your trunk later. Yeah. But yeah. You know, well, later, they probably apprehended you by then, right? So I feel like it's only long enough for you to be able to stop quickly and pull it off. But that's where I think the glue probably keeps it from being easy to pull it off. I mean, on balance, it sounds like of the tech innovations that could end humanity, I'm mostly fine with this one. And as someone in the chat says, you can misuse anything, many things in particular that police people, policemen have police officers have can be misused. And it sounds like a lot of bodies are weighing in on when you can use it when you can. And there's similar legislation on GPS tracking and other trackers you might put in people's cars. So if on average, we're we're mostly minimizing people dying in high speed crash or high speed car chases and we can tackle the other surveillance issues that come up with not just this, but other things then plus we get to live in Batman world. It's kind of awesome. Exactly. Exactly. I like to find it mostly fun. Frankly, the misuse of this is probably not firing a dart at people, which is more likely to be seen. It's like secretly hiding the tracker in the wheel well when no one's looking, right? So I'm less worried about the misuse of them firing foam darts at wild at people. And yeah, and I think this is, you know, a decent way to minimize the risk to innocent bystanders. So even here in New York City, where we're not known for our car chases, we have times where police are really screaming down Sixth Avenue and. Yeah, right. So why not? Yeah. Have either one of you thought about the Roman Empire today? I think you're going to. No, no. Yeah, not today. They used to have high speed chariot chases in the Roman Empire. I don't think about the Roman Empire all that often, but that didn't stop Roger and I from putting together a top five ancient Roman technologies we still use today. If you would like to catch up on what from the Roman Empire is still used in the technological world right now, you can catch it in our top five at Daily Tech News Show on Tiktok, DTS picks on Instagram and YouTube.com slash Daily Tech News Show. The New York Times published a story last Friday regarding research showing that China is producing 50 percent of the world's talent in A.I. and the United States produces 18 percent of the global total of talent in A.I. For China, that number is up from a third from three years earlier, whereas the U.S. number is consistent. We just keep pumping them out at an 18 percent rate. The study came from Marco Polo, a think tank run by the Paulson Institute that promotes constructive U.S. China ties and was based on the backgrounds of researchers who had their papers published at the conference on neural information processing systems in 2022. Now, Andrea, you taught data science in the U.S. and in China. So you you have a unique perspective on this. Do you trust this metric? And and what do you make of this disparity? Yeah, so it's it's two good questions. And and I did teach in both China and the U.S. And in both contexts had students from China and from the U.S. and then in the U.S. from China from the U.S. so that all four quadrants of my two have been covered in this. And I do think that, you know, two things stand out to me. One, I was a little bit disappointed to find when I went to China to teach that all the stereotypes are true. The Americans were terrible at math. The Chinese students were amazing at math and we had a real crisis when it came to like the placement tests because they were supposed to be these integrated courses and all the students who had come up in the Chinese system were just placed much higher. But then we had to teach essay courses and all the American students could write long essays about this is my opinion on that. This is my opinion on that. And my Chinese students would really struggle to write, you know, and so I was really dismayed that all these stereotypes came true. We worked a lot to undo them, but there is truth to it. And so at first plus, first plus you read this article and you say, well, it's not surprising, right? China has a long history of really outstanding math, engineering, STEM programs. Traditionally, they do send many, many students to the United States, many of whom I've taught at the undergraduate level. And I also do admissions for the graduate level. We have many, many excellent applicants from China. Many students in the NYU master's program in data science are currently from China, many from India as well. But China is the largest group. And the training is excellent. They're excellent students. Again, I'm speaking in broad, broad general terms about the education system, where I get a little bit uncomfortable with this article and with the Marco Polo study and with the broader conversation around kind of the AI talent competition. One, I don't know that we need to make it this geopolitical contest. Marco Polo is really designed to, as you said, promote cross national cooperation. So this sort of like how many of us are from here and how many of us from there doesn't feel in the spirit of that. But much, much more importantly, I completely disagree with how they're deciding what counts as AI top talent. So they're using the term both in the article and in the study that they're referencing talent as this like I think the study even says it's very quantifiable. So no problem there. I don't know about you, but I don't know how you quantify talent. I work in education. And one of the things I work on in data science is turning things like talent into numbers, like how companies conduct performance reviews, whether you can use AI for that. And turning talent into a number that you can measure is very, very difficult. Can't not to say it can't be done, but it's very difficult. And I've yet to see anyone do it right. And this particular metric. Oh, what was the national origin of the people who were co authors on this study that was accepted into this particular conference? Sure, that's one way of thinking about people who are maybe sort of big movers or or highly influential in the AI space. But even this chart that you're showing now, it says, our top tier AI researchers, that is such a broad term. And we're really focusing just on a cohort of papers that were accepted to one conference. First of all, there's a lot going in AI that is not just that particular conference. And the article does do a good job of distinguishing between training and generative AI related tools and other forms of more traditional AI. But either way, it's such a narrow, tiny little sliver. There are a million ways that we could talk about contributions in AI. And that brings me now I'm really on my soapbox to my broader point, which is I struggle with this with my own students in data science. There's this idea that in order to contribute to AI, you have to study programming and only programming. And if you even stray so much as to like too many statistics classes, you're somehow diluting away from this this march towards the the programming utopia that we're all seeking as top tier AI talent. When really a lot of the big innovations that happen in AI, they happen since Alan Turing's time till now, happened because people have ideas that come from other fields. And so I don't want to sound like the old person yelling about liberal arts. But my students are the most creative projects who do the best senior research projects who do the most interesting things in my biased opinion after graduation, have skills in other areas. And my students in my teacher senior course in natural language processing, I'll end my rant here. The students who can program really, really well are valuable and they're they do good work, but they aren't able to actually ask and answer research questions. So the actual act here we go of turning something interesting, complex, important in the world, like language or translation or, you know, reading resumes or whatever it is that you want to do with AI, you need to have these skills in these other areas. This doesn't mean force everyone to take a poetry class and call it good. But what it means is that being a top tier AI talent, isn't just did you contribute to this one tiny conference in this one niche field? Or are you the greatest programmer who's ever lived? And frankly, last last real last thing, I work with a lot of employers who want to hire AI talent, data scientists, computer engineers, software engineers, anyone, and they're just looking for the list of Python programs that they speak or the languages that they speak and they have no concept of the importance of people who can think about ethical implications of the work that they're doing. Think about Elon Musk and the changes that have taken place with X getting rid of a lot of different teams that aren't necessarily programming forward. It has a real impact on the AI ecosystem that we're living in. And I really, really have a problem with thinking that you have to be a very narrow type of computer scientist in order to be a valuable piece of talent in AI. I guess the only. Yes, thank you. No, that was very well laid out. I didn't mean to. I actually think it's great. No, it's good. I mean, hey, you know, not everybody is passionate about this. So it's really good to hear from somebody who's like, this is how this affects me and a lot of people that I'm educating. Yeah, no, I have students who are who they narrow themselves down. They come to me and they say, Hi, I'm interested in being a data science major going to grad school for data science working in tech or whatever. And I am also interested in music. I'm a classical piano player. I'm also interested in creative writing. I'm also interested in biology, biochemistry. And they say, but if I want to get a job in AI, I need to really zoom in. So I'm not going to take any of those classes and I'm only going to take data science courses and our data science courses are excellent. But you're doing yourself a disservice, not letting yourself be exposed to other ideas, other questions, other ways of thinking that then you can combine with your data science skills. That's where innovation is going to come from. So we've just got this tunnel vision and whether they're coming from the US or China, who cares? Right? It's it's what's the breadth of knowledge that you have in addition to the skills and in this arms race is going to drive us into the ground. I think I think that the worry is that if China is producing 50% of even the the narrow talent that they'll they'll end up having more people able to make advances faster than the rest of the world and keep them to themselves, right? And not share them with the rest of the world. Do you like that? Just just to wrap this up, you know, how concerned are you about that? So there's two two views on this, right? So one thought I have as I read all of this and follow along is you know, Soviet space race. I'm watching for all mankind right now US versus Soviet space race. And so there is this kind of talent arms race that could be going on or could continue to go on between China and the US. And look, I left the United States to go get a job in China. So it's definitely not the case that it's we can assume now as Americans that all the top talent, top talent in AI are going to want to flock to the US to work in US based companies. There's a really great exciting things. One good thing is that this kind of competition could spur innovation. So that could actually be a really good thing for the world of AI. But a concern, of course, as you said, is we can get into a world where we're weaponizing these things where we have cyber attacks or we have, you know, other sorts of unhealthy competition that could be going on. But the last thing I'll say about this that that comes to mind for me is I was in China for three years. I was at New York University Shanghai campus. And I was there when it first started. So this is 2013 to 2016. And one of the reasons the Chinese government wanted NYU to open its branch in China as opposed to, you know, let it go somewhere else, was that the Chinese government at the time, so we were told, was very concerned about the lack of creativity coming from their students. The students were very excellent. They were taking, you know, nailing tests in general. China is very good at replicating existing technologies and speaking in broad terms. But the Chinese government was very eager to specifically bring in liberal arts thinking and bring in creativity because they were lacking in innovation. So even though more of these, you know, if you take the top tier talent measure as correct, even then I think we're missing a big piece, which is what do you do with that talent? Do you have the corporate organizational structures to breed innovation? Do you have people who are asking questions and challenging hierarchy? Do you have a space where people even think outside the box? And again, I'm speaking in broad terms, but my students in my course here in the United States who were from China on average struggled most with homework assignments where I said, OK, here's a data set. Here's a couple of interesting questions to explore. What method would you use? What are the shortcomings of using this method? What if you did unsupervised learning? What did you learn about the model? These open-ended questions students would really struggle with. And so there's a piece of creativity that for all the problems in US education, I think we're still doing very, very well. And I have yet to see that that has meaningfully changed in China to the point where the Chinese government is admitting they were like, can you offer a course on creativity? And I was like, I really think that's how it works. But yeah, sure. Yeah, we we demand that you be creative. Yeah, it goes down over there. Yes, to creativity. Yeah. Yeah. All right. Before we wrap up, let's check out the mailbag. Let's do it on Tom's latest editor's desk. He dealt with a question on how to keep a small website afloat in a world where AI tools are summarizing content without sending folks to a website itself. Dan wrote in who works in digital marketing. Hi, Dan. And responded with some thoughts, including we manage Dan's company, a lot of paid search ads for our clients for our for our ad clients. And this is this very conversation has been discussed in our teams ultimately, Google being the giants of paid search are not going to give up that revenue easily as they're both building LLM platforms. I believe paid search will show up in LLM conversations in the future. And I can't wait for the Atlantic think piece on that. Oh, yeah. No, but but but but thank you to thank you to Dan. We got a few responses about this. Keep keep that feedback coming feedback at daily tech news show dot com. Andrea Jones, Roy, thank you so much for being with us on the show. Let folks know where they can keep up with all of your musings. You can definitely keep up with all my musings at Jones Roy on all the social medias J O N E S R O Y as you can see. And a brand new website, data science needs you dot com where you can check out. There we go. It's a very basic website, but I'm producing videos from my data science for everyone course at NYU specifically to attract folks who think data science and programming and STEM and blah, blah, blah are not for them. And there's a free online textbook, so data science needs you dot com. Patrons, stick around for the extended show Good Day Internet. It's Friday. And Andrea, among many other things, is also a comedian. So we're going to play a few rounds of who am I and try to guess which tech nerds became famous comedians besides Andrea. So stick around if you're a patron. I'm famous, but OK. We're getting into wait, wait, don't tell me territory and I love it. But just a reminder, you can catch our show DTNS is live Monday through Friday at 4 p.m. Eastern 20 hundred UTC. You can find out more at daily tech news show dot com slash live. We hope you all have a wonderful weekend. We're back on Monday with Justin Robert Young joining us. Talk to you then. This week's episodes of Daily Tech News Show were created by the following people, host producer and writer Tom Merritt, host producer and writer Sarah Lane, executive producer and Booker Roger Chang, producer, writer and co-host Rob Dunwood, video producer and Twitch producer Joe Coontz, technical producer Anthony Lemos, Spanish language host writer and producer Dan Campos, science correspondent Dr. Nikki Ackermans, social media producer and moderator Zoe Dutterding, our mods, beatmaster W. Scottis one bio cow Captain Kipper, Steve Guadirama, Paul Reese, Matthew J. Stevens, AKA Gadget Virtuoso and J.D. Galloway. Modern video hosting by Dan Christensen, music and art provided by Martin Bell, Dan Looters, Mustafa A, Acast and Len Peralta. Acast ad support from Tatiana Matias, Patreon support from Tom McNeil. Contributors for this week's shows include Chris Christensen, Scott Johnson and Justin Robert Young. Guests on this week's shows included Trisha Hershberger, Andrew Main and Andrea Jones Roy. And thanks to all the patrons who make the show possible. This show is part of the Frog Pants Network. Get more at frogpants.com. Hope you have enjoyed this program.