 This 10th year of Daily Tech News show is made possible by its listeners. Thanks to every single one of you, including John and Becky Johnston, Chris Benito, Steve Aderola, and our brand new patron, Keith. Woo, welcome, Keith. On this episode of DTNS, Microsoft wanted Apple to buy Bing kind of how to keep your website from being used to train AI and the countries who stand to gain the most from people using those AI tools. This is the Daily Tech News for Friday, September 29th, 2023 in Los Angeles, I'm Tom Merritt. And from Studio Sushi, I'm Sarah Lane. On the show's producer, Roger Chang. And joining us, the producer and co-host at Momentus TV and AI named the show, Tristan Jutra, welcome. Happy to be back. It's good to have you. How's the new show going? Aside from its impact on my sleep schedule, well, well, we're doing it in the middle of the night. Isn't that always the way like it's going well and I haven't slept? Exactly, yeah. This episode six just dropped today, our AI clones coming for your job. Ah, find out on AI named the show. Meanwhile, right here, we shall start with the quick hits. The Wall Street Journal sources say Chinese officials told Apple that it must strictly implement rules to ban unregistered foreign apps, with sources citing recent meetings between officials and Apple staff. China restricting the apps would prevent Chinese iPhone users from downloading some Western social media apps like Instagram, X, Facebook, YouTube and WhatsApp. You don't often hear of China easing restrictions on the internet, do you? But the cyberspace administration of China has proposed loosening cross-border data security controls in order to ease the concerns of foreign businesses. These are the rules that govern how companies are allowed to move customer data between countries. So in this case, between China and data centers outside of China, the proposal would no longer require specific consent to move data that doesn't contain personal or other important info. Now, how important data is defined is going to be important. Right now it's still a little vague, but it is expected to be defined in the final version. France's competition authority rated NVIDIA's local offices this week due to supposedly suspecting that the company is engaging in anti-competitive practices. Now, the regulator did not elaborate on which practices it was specifically investigating or even which company it had targeted NVIDIA or something else, beyond that it was in the graphics card sector. The Wall Street Journal sources say that the operation had indeed targeted NVIDIA, the world's largest maker of graphics and AI chips. US Supreme Court announced Friday it will hear two cases about laws in Florida and Texas that would restrict how companies moderate content on social media platforms. We've talked about these before on DTNS. It's been a while though, because they've been working their way through the system. The two laws would prevent companies from demoting or removing content from their platforms on the basis of viewpoint. It would also require transparency in moderation rules and require companies to stick to those rules. There's a few differences between the Florida version and the Texas version, but that's the overall gist of both of them. Both laws have been prevented from going into effect while the legal challenges have worked their way through the courts and the Supreme Court will hear arguments on the cases likely sometime next year. Few of us this morning might have heard friends of ours who use Discord saying, is something wrong with Discord? By the time I woke up, apparently it had been put to rest, but kind of a big deal. Discord messaging platform used by gamers, livestreamers, general chatters, people who do things like what we do on DTNS appeared to be down early Friday for many users. In a statement, the company said, as an explanation, we're experiencing unusual traffic spikes that lead to users being temporarily blocked and we're working on mitigating that issue. The issues appear, at least at this point, to be related to Cloudflare, which was having problems with its dashboard and API service and is undergoing scheduled maintenance. More fun facts, keep pouring out of the USFTC's antitrust case against Google, Microsoft Business Development Vice President John Tinter testified Thursday that Microsoft CEO Sacha Nadella met with Apple CEO Tim Cook in 2016 to discuss investing billions, multi-billions of dollars in a deal to replace Google as the search engine on iOS. Bing, at the time, powered Siri and Spotlight searches, it did that between 2013 and 2017, but Microsoft at the time wanted to expand that to Safari, where Google was still used and beyond. In 2018, after their deal had expired and Google was now on everything, Apple and Microsoft met again to discuss a search deal outside the United States. Well, you don't want us in the US, how about if we do it elsewhere? Microsoft was also apparently talking to Samsung about making Bing the default on its platforms and Bloomberg sources, this wasn't part of the open court testimony, but Bloomberg sources report that in 2020, Microsoft pitched Eddie Q at Apple on the idea of selling Bing to Apple, and that's the thing making all the big headlines today. Apple apparently never took any of these deals, mostly using Bing's existence as a bargaining chip to get more money out of Google. Samsung never seriously entertained the ideas and apparently through back channels, politely asked Microsoft to stop pitching them, saying that their president to Pro Tem was too polite to tell them to stop, and Samsung did that because they didn't want to threaten their much larger relationship with Google, which involved all of Android, not just search. So the big headline here is Microsoft pitching the idea of selling Bing. That clearly never was serious. I don't know, Tristan, should any of this matter to any of us as consumers? Well, it's interesting in some of the classes I taught in a past life about things like social media and search engine optimization, I would often ask people, it's like, why do you think when you open up your iPhone, for example, and you go to search that it goes to Google? And how did that arrangement happen? Or when you, in that period that you mentioned, when you're asking Siri a question and it goes to show you web search results sometimes we asked to search, it would give you Bing results. And a lot of regular folks had never really thought about that. And so he said, it's pay to play. And estimates for that time period for Google anyhow, have ranged anywhere from $8 billion to $18 billion. It's never actually been disclosed, but that's a lot of money that Apple is not that interested in letting go due to a regulatory interference, but then the whole notion of Bing trying to sell itself to Apple, Microsoft divesting of that, which kind of runs contrary to Apple's build it here, not invented here syndrome, they do largely smaller acquisitions. For example, 21 AI companies in the last couple of years, but Beats, I think it was their biggest acquisition in like a long, long time. So to go and buy something like Bing from Microsoft would be kind of out of character. Furthermore, people have noticed over the last couple of years an Apple bot actually crawling people's websites as well, leading to speculation that maybe Apple was creating and designing their own search engine in-house, which would be a pretty heavy lift to try to make something competitive with Google, which has become so entrenched over the last couple of decades. Perhaps this Apple bot was using to train their large language model, which again is rumored to make, maybe someday make Siri smart, finally. Well, you know, hearkening back to 2023, I remember Siri was the first AI powered assistant that I was using and I remember being in the car and trying to just like more or less, make her say something crazy, like, oh, let's figure out how to make the bot not say the thing that you want to say. And that's how we know that humans are still in power, but it was a Bing thing. I had kind of forgotten about that. That was 10 years ago and it was Bing powered search for Siri in the early days. And that was at the time where a lot of us were kind of like, this is cool potentially, doesn't work that well, but like potentially pretty cool. So I can see where the two companies would have had many conversations over the years to be like, okay, how do we make this better? How does this partnership make sense? And I guess it didn't. The conventional wisdom in all the stories I read today was that Apple was playing along Microsoft in order to drive up the price with Google. They were happy to have Bing around because it meant they could ask Google for more because they had a way to walk out of the deals. My theory on Microsoft having a low level exec or lower level exec, right? We're not talking Satya Nadella at this point where somebody else coming in the room and saying, hey, would you want to buy Bing? I feel like they didn't expect that Apple would want to buy Bing. They didn't really want to sell Bing, but they figured that might be the only way to get their attention and maybe get into a room with Tim Cook again. Because it feels like at that point, Apple had just said, you know what? Sorry, we just don't want to buy your thing. It's similar to Samsung having to have employees through a back channel say, our boss is really tired of you asking about Bing. It's just not gonna happen. Can you stop? So when they're discussing pay to play with Eddie Q and then they go, hey, maybe you just want to buy us. And Eddie's like, I got to talk to my manager. Yeah, right, exactly. That's what they were hoping and that Tim, he would then pull Tim Cook out of his office maybe. I don't know. That's just my conspiracy theory. I gotta admit, I've been trying to use DuckDuckGo for the last few years and it's powered by Bing with some extra Zhuzhan top and I still find myself going to Google quite often because it's different, but it's still not quite there for certain things. And then of course, a lot of us are using Bing Chat. I think that was the genius move with all of this because one of the theories that people have as well is that maybe Microsoft didn't want to pay the kind of money that Google was paying for the exclusive default for the Safari browser on iOS and even hang on to the Siri default either because they could use that money for other things, maybe redirecting it to OpenAI, for example, they save their pennies. So they get their $11 billion into OpenAI so far. So maybe that was the smarter play because so much more interest in Bing since early this year after the partnership and they've been rolling the GPT functionality into Bing Chat. I know I've been using it a lot more since then. Yeah, that's a smart thought. Speaking of AI, London's Capital Economics issued an assessment of which countries will benefit the most from new technologies marketed as AI. The study looked at countries potential for innovation, use of AI, adaptability to its effects. The US topped the chart, followed by Singapore, followed by the UK. All three countries have successfully attracted top talent in the field and have favorable business policies. The UK in particular benefits from its higher education system, Google's DeepMind based in London, Switzerland ranked fourth, Sweden ranked fifth. The two countries ranked first and second in the world in adaptability with a good track record of redeploying resources in the face of new technologies. China ranked relatively low because of regulatory barriers and government intervention in the private sector. We kind of talked about that a little bit earlier in the show, but yeah, what do we think about this? Tristan, you're covering this beat regularly. Sure, yes. International course. I'm sorry we don't have the rank of Canada on here. It just wasn't in the story we looked at. I know. Where the heck is Canada? So do you think it's a coincidence that three of the top five are in Europe, but two of which aren't part of the EU and none of which are part of the Eurozone? That was one of the conclusions of the report is that a lot of Europe lacks the cloud infrastructure needed for AI processing and therefore has to go elsewhere to do that. The financial backing isn't there everywhere. It's there in the UK. It's there in Switzerland, but it's not in other places. And most of all, the regulatory rules are very strict in Europe. So that's a good question, Tristan, which is the European Union proper has done a lot of work to protect jobs, protect privacy ahead of the implementation of AI, but the implication of this report is that in doing so, it has reduced the competitiveness of AI within Europe. And don't get me wrong. I'm from Canada, so we love us some regulation up here, but when I see the stuff that the EU's been doing over the last number of years, it makes me feel like Peter Thiel. Like, I can scream you with a chair. Oh, gosh. So... Sorry for that. We're all seized heading now, apparently. So when you see the kinds of moves that the EU has been making, well, we have the today's story about France, part of the EU, being aggressive against Nvidia, being aggressive against Activision Blizzard acquisitions, being aggressive against Apple with the whole USB-C thing that they were gonna get to anyway, like it was on most of their other devices. It's just this one after another after another, and you can't help but think that that's eventually gonna maybe create a bit of a hostile or at least a chilly climate for investment, because one of the other stories that's wrapped up with this is recruitment. And if you have lack of funding because of concerns about over-regulation, then it's gonna be hard to recruit. And then if it's hard to recruit, it's gonna be hard to get funding. So you get this vicious cycle where they're creating this not super venture capital friendly environment for tech, and Europe's gotta do something, right? To get going again. And we've got these, you know, UK, Switzerland and Sweden, which are kind of, you know, they're Europe, but they're kind of not Europe, each in their own way. And so they're not sub, at least UK and Switzerland, don't buy into a lot of the regulation from the super government, the EU parliament, which, you know, it's like, they just red tape. I mean, I know Switzerland doesn't, but Sweden does not as well. Sweden is part of the EU, but they're not part of the Eurozone. So they don't use the, they're not currently using the Euro yet. And so there's a number of things that would be associated with that. So they probably have a little more wiggle room than those that are like fully bought in. And of course, you know, the Swiss still on the Swiss Frank, UK is on the pound and they're not part of the EU anyway, either of them. Yeah. And this is one situation where the UK leaving the European Union has freed them up a little bit and provided a positive in that they, it may be harder for them to collaborate across the channel with European institutions. That's one of the downsides, but it's got some flexibility that seems to be at least in the estimation of London's capital economics, which granted is in the UK in their estimation, it gives the UK a potential advantage here. But at the same time, the things that the EU are doing, privacy is a good thing, right? You know, promoting competition is a good thing. You know, cracking down on anti-competitive practices and the US is generally a bit more liberal when it comes to these issues. You know, the FTC's latest adventures the last couple of years, not withstanding, but you know, it's all these things happening. There's a reason why the US attracts so much money and so much talent and generates so much innovation. And we're not seeing quite as much of that out of places like the EU. And not to pick on Europe, that's one of the reasons they dinged China on this as well, was the heavy regulatory practices and the uncertainty of, well, if I start a business there, am I gonna be able to keep that business there? I have to partner for things like data storage within the country, which is why that loosening of data transfers earlier, I think, was fairly significant. It's China's, the first indication recently that China's recognizing that this is really a break on foreign companies coming in and that's hurting their economy. Yeah, and the EU's looking at China and thinking, life goals. I don't know if it's that intense or not, but maybe, maybe it is. Well, folks, we do a lot of other stuff on our channels. The patrons get a lot of cool shows if you've been enjoying Free Preview Week, you know that. One thing that's available to everybody, not restricted to patrons is Tom's Top 5. It's a show I do on our YouTube channel where I break down five things to know usually about technology. Sometimes I stray outside of technology and have a little fun. This week's episode is the top five greatest video game characters of all time. If you want to know what made the list, well, you got to find out by going to youtube.com slash daily tech news show. Well, Google announced support for an extended flag in websites, robots.txt file. You know it, you love it. The let's exclude its site's public data from being used to train Google's AI models. Tom, what's up? Yeah, so if you don't know or love yet the robots.txt file, it's a file that sits in the root directory of a website. So it's there, the first directory that a web crawler that's going around the internet, looking at everything that's on the internet, will see. It is most often used to tell those crawlers what parts of a site it can index and what it can. There might be a lot of reasons where a site says, we for search engine optimization or even just for limited privacy reasons, we don't want this part of our site to show up in a search engine. So they'll put an entry in robots.txt that says, hey, Googlebot, which is the name of the crawler, don't access anything below the homepage, let's say. Google has now added support for a token called Google dash extended. So in your little txt file, you can put instructions for Google extended and when a website includes that, it restricts the use of data collected by the web crawler from being used in AI training without restricting it from being indexed for search. Before it was kind of either or, if you wanted to restrict it from training, you had to restrict it from search. Now you don't have to. OpenAI provides a similar option. Tristan, what do you think of this move? Well, I think it's good to give people options and opt out is great. Some people are going, well, why not make it opt in? And because if they did, you know, defaults matter, right? So if they did, they would barely have anything to train on because most people would just not get around to it. Even people who wanted to do it wouldn't bother to go put it in. Exactly. Now I have to admit I have a strange affection for robots.txt files. Again, from my SEO days and SEO instructing days because it's this weird little text file that a lot of people don't even realize exists. And you can go... Well, and like just explain, like why is it, you know, something that is so prevalent today? Well, if you can go to any regular, any of one of your favorite websites and after the domain name slash robots.txt and you can have a look what they've got in there. And it's actually very telling what search engines or sorry, what websites would like the search engines to A, crawl and B, index, which are two different things. And from a search engine optimization point of view, there are certain things that you don't want the search engines to index because it can actually muddy your SEO profile. If you wanna be known for certain things only and ranks very well for certain things, then it's actually in your interests to block certain pages from the search engines. So the search engines have a clear idea of what your site is about and what the clusters of pages are about. So it's actually really interesting to take a look. Say if you go to apple.com slash robots.txt, you can see instructions not only to the Googlebot but various other spiders and or crawlers and see what they are actually blocking off a lot of their e-commerce pages, for example. They don't want those pages cached for one things because they may show obsolete pricing. So a lot of websites will do that for promo pages, landing pages, splash pages, things like that. So that potentially obsolete content doesn't show up in search results. And that's, I've actually found deals that way going through a search, but then you get to the deals page and realize it's expired. So that's not necessarily a great look for websites either, but you can also see the kinds of things that Apple, for example, is blocking from some of the Chinese search engines. I was just noticing that because I'm looking at that since you mentioned it. Including product red. And because Apple doesn't have product red in China and red means something different in China than it does for product red globally for the fundraising to fight new to combat HIV around the world and particularly Africa. That's not something that China is not, that is necessarily on board for, right? So they may have red Apple products in China but they're not product red. So this is whole sort of curating the visible web pages for not only the search engines but the search engine results what people are actually gonna see. And then just very quickly, the difference between the crawling and indexing is that allowing them to crawl lets them, the spiders of the bots follow the various links through your site to go through the entire hierarchy and get an idea of the topography of your website. And you can prevent that, but you can also alternatively ask for certain pages not to be indexed. They can be crawled, like all the links can be followed but not necessarily indexed. And when they're not indexed the snippets won't show up in the search engines and the pages won't be cached as well. So there's different reasons to choose one or the other or both. Yeah. The robots TXT on the DTNS site just says don't index or crawl our admin section. Thanks. Everything else is fine. One post-secondary institution I was working for they were blocking a crawling of a folder called Fluffy Bunnies. It's like, what is in the Fluffy Bunnies folder? Interesting. One wonders. Sorry, go ahead. Oh, I was just gonna say, on the subject of this, Google also announced a few new features on its search experience, SGE including large language model features in Google search now available to ages 13 through 17 if you're a human in the U.S. Previously restricted from those accounts. They'll still add a way to add it through search labs and the company says it's also adding stronger protections for outputs related to illegal or age-gated substances or bullying. Google is also adding an about this result note to AI generated responses from SGE. It will give more context about how the response was generated. And Google also updating its model to detect false or offensive premise queries and respond with more accurate responses as a result. It's also working on using LLMs to analyze its own first draft responses and rewrite them for higher quality and safety. Starting to seem like Google and other AI tool makers are kinda worried about safety because that's what happens when you're big enough. Is that good, Tristan? What I wanna know is when they're gonna open it up to 52 year old Canadians, because we still can't access a lot of- Because you can't get it in Canada yet. No, no Google bar, none of this stuff. So that's why we're team being here in Canada. Let me add it. Yeah. Exactly. So I guess the question is this, how effective will this be? Or will this just be a challenge for enterprising teenagers to get into these tools regardless of the age? I mean, how many kids do you know that got onto Instagram or Facebook before they were 13? It's really not that complicated. They've always got a friend that can get them in. What have you? Now, I think it is key that they are trying to dissuade the AI from responding to prompts regarding bullying or the consumption or use of either illegal substances or age restricted substances. And maybe I'll know what those are. And, you know, but those who want that information are gonna find a name. I mean, Google search still exists. Right. It seems like people are worried more about the safety of the barred prompts than they are about the rest of search. And- That's what's getting all the heat right now. And it's gonna limit the effectiveness of barred, which again, that's the trade-off. And many of you may be fine with that trade-off, but it does seem like Google is spending a lot more time telling you what it's doing to stop the chatbot from saying anything possibly offensive than it is telling you, oh, we've made it better. And even if it won't tell you how to make a Molotov cocktail, you can still find that on Google. Yeah. Yeah. Well, also things that you can find on Google, I suppose, officially Netflix has officially ended its DVD service. Pour out a little liquor, everybody. Netflix launched in 1998. And by the way, with the company's announcement of winding down DVD services, reminded folks, Beetlejuice, the movie, the first title shipped. The company set a record in 2011 with 4.9 million DVDs shipped in a single day. In total, the company says it shipped 5 billion DVDs. This is a momentous day. I know, I know. You know, I mean, besides Roger Chang, I don't know who else cares all that much. I got my last three discs the day before yesterday. And you get to keep them, right? I get to keep them. I mean, they will take, they will accept returns until the end of the next month. If you don't want to keep them. If these small little keepsakes are taking too much space in your home. But, you know, I've always wanted a copy of Shortcut. What'd you get? You got to tell us the titles. Everybody wants to know what you got. I got Shortcut, I got Shortcut's, you know, Robert Halton movie. Yes, yes. And then I also got volume one and two of the new adventures of Batman, The Complete Series, which is the 1960s animated Batman, featuring Bert Ward. It's still doing Robin, but someone else doing Batman. They held it. Are you going to file those with your late blockbuster returns? I will file it under the other things I have in my room that I will probably forget in a month's time. And if a Canadian says quickster three times quickly, we finally get the DVD by mail service here, because we never had Netflix by mail here. We had a copycat called zip.ca. But we just got Netflix purely streaming, I mean, which was great, although it was like a quarter or 20% of the catalog. I don't know if they'll mail to Vancouver, but there's a video store in Seattle that's doing by mail now, because Netflix is getting out of the business. So it might be worth looking into. You fill that gap. Just like Redbox, is Redbox still a thing? Yeah, there's one, my local grocery store. I saw some guys stand in front of it. Well, trying to pick a movie the other day. Amazing. Well, Trish and Jutra, thank you so much for being with us. Whether you have DVDs or not, we really, really appreciate having you with us. Let folks know where they can keep up with your work. You can find our regular live streams at 7 PM Pacific Time at momentus.tv. And Teja from Momentus and I started a little show thanks to the encouragement and support from all of y'all at DTNS and that's called AI Named This Show, which is, you know, we aimed for the dumbest possible name, but at least it's memorable and it's got great SEO. So you can find us at AINamedThisShow.com. And I'm at Trish and Jutra, and that's J-U-T-R-A-S on pretty much all the socials. Excellent. Well, folks, it's the last day of free preview week. We do have a little something coming into your feeds tomorrow on Saturday. All this week, we were giving everyone access to the Good Day Internet Extended Show. I hope you've enjoyed that. If you did and you want to keep it coming, you can just become a patron. Go to patreon.com slash DTNS. And all of you all stick around for the Extended Show Good Day Internet. It's Friday, so we're playing another round of Who Am I? Roger has researched great tech people in history and we're going to try to guess who he picked one clue at a time. Come play along with us. Do so. Just a reminder, we do this show live. Monday through Friday, 4 p.m. Eastern, 200 UTC. You can find out more at dailytechnewshow.com slash live. We will be back on Monday with Justin Robert Young joining us. Have a great weekend, everyone. 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 Coons, technical producer Anthony Lemos, Spanish language host, writer and producer Dan Campos, science correspondent Dr. Nicky Ackermans, social media producer and moderator Zoe Detterding. Our mods, Beatmaster, W. Scottus1, BioCow, Captain Gipper, Steve Guadarrama, Paul Reese, Matthew J. Stevens, aka Gadget Virtuoso and J.D. Galloway. Mod and video hosting by Dan Christensen, music and art provided by Martin Bell, Dan Looters at Musaffa A, ACAST and Len Peralta. ACAST adds support from Tatiana Matias, Patreon support from Tom McNeil. Contributors for this week's shows include Chris Ashley, Scott Johnson and Chris Christensen. Our guest this week was Tristan Jutra 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. Diamond Club hopes you have enjoyed this program.