Tech News Weekly 383 Transcript

Please be advised this transcript is AI-generated and may not be word for word. Time codes refer to the approximate times in the ad-supported version of the show.

0:00:00 - Mikah Sargent
Coming up on Tech News Weekly. Dan Moren of Six Colors is here. We kick off the show by talking about the DOJ winning in its case against Google and its ad tech. Then OpenAI rumored to be building a social network. After that, mark Zuckerberg just couldn't make it happen with the FTC and its own meta antitrust case. Before we round things out about ChatGPT being used for reverse location search, it's GeoGuessing. All of that coming up on Tech News Weekly.

This is Tech News Weekly episode 383, with Dan Moren and me, Mikah Sargent, recorded Thursday April 17th 2025. Google loses DOJ ad tech case. Hello and welcome to Tech News Weekly, the show where every week, we talk to and about the people making and breaking the tech news. I am your host, Mikah Sargent, and today I am joined by a special guest co-host. It is Dan Moren of Six Colors and Clockwise and other podcasts and activities. Hello, dan.

0:01:22 - Dan Moren
Yes, I do many activities, Mikah. Thank you for remembering all the things I do. It's a long list Actively.

0:01:32 - Mikah Sargent
It is good to have you here on the show. For people who are tuning in for the first time, we typically have the start of the show where we have our stories of the week, and then, throughout the show, we'll have on some guests who join us to talk about the things they're writing about or that they are covering or have created this week. It is a story of the week episode. I'm pleased to say, dan will be joining us for the whole episode. So, without further ado, let's kick things off, dan, with your story of the week.

0:02:07 - Dan Moren
Well, this was kind of late breaking, I think, as we started putting this together. The US Department of Justice has won its antitrust case against Google. This is specifically now just to clarify. There are a couple of different things. They recently also had a loss over search and they're having a monopoly in search. This is about advertising and Google's role in the advertising market. What they basically established was that Google has been anti-competitive in the way that it's established and maintained its monopoly power as a publisher for ads and in creating ad exchange markets, and basically the way they tie those things together, both technologically and contractually, lets it hold on to that monopoly power in those markets. So it's also important to understand again, this is technical, this is legal stuff. Either Mikah nor I are lawyers. Last I checked, you're not a lawyer, right, if you?

lawyer right you have to tell me that's the law, but this is so. One thing is not included here is that they were dismissed. The charges that Google operated monopoly and ad networks yeah, so this is it's slicing a little bit fine here.

And of course, Google has said they're going to be appealing this decision Unsurprisingly. We'll see where that goes. But specifically, Google says that the acquisition of things like DoubleClick, which they famously bought as sort of a big competitor, did not harm competition. And so they argue that people use Google ad technology because it's the best right, it's the cheapest, it's the most effective, et cetera. You could also argue that's the case because there's not a lot of competition again.

So what does this mean, I guess, is the question. Well, we don't, I think, yet have a you know sort of what the remedy would be here. I'm not sure that that has been made public yet and, of course, if Google is appealing, that will be pending anyways. Right, that's not going to happen immediately, but it is significant because these are two back-to-back cases in which Google has taken a blow over its role in markets where it is a dominant player, even if it may or may not think of itself as a monopoly. There's no arguing that it's dominant in search and it's dominant in ads. I mean, those are its two biggest markets. So whatever the remedies may be, they are likely to be, if not far reaching, then at least impactful in terms of how they change Google's business. It's going to have to make some changes and adapt to figure out how it's going to comply with whatever remedies the court sets down in this case, so that could be presumably a large shift in some of these particular types of markets.

0:05:04 - Mikah Sargent
It's so confusing. I mean, obviously this is a giant corporation and has done some, you know, some questionable things. Still, for the people who are involved in this and who have to figure this out and break this all down, I can't imagine. I mean, because, as you point out, there's sort of these two areas the ad network versus the ad uh, like publishing, yeah, yeah, the publishing. And so when it's almost like with that, they're saying we're not going to get you twice, right, you do both. And it's the part of the problem is that you do do both. We're not going to get you twice, right, you do both. And it's the part of the problem is that you do do both. But we're going to focus on the one, because if you you're able to use the company to buy ads and then the company also, uh, then serves those ads, it, it's there and that's just very again, I keep, I know I keep saying this, but it's incredibly complicated and so I think, on both ends of this, I'm very impressed with anything that can come with it.

It's a lot to pull apart, and that's why I think, too, google has so many things going on right now in the antitrust space, including its search, including the browser stuff. Obviously, there's much with these big corporations.

0:06:31 - Dan Moren
Right. I mean there's a lot of slicing this thin right. We saw, under at least the previous US administration, a willingness to take on a lot of these big tech. We're going to talk about another one a little bit later. But Google Google we also saw. Remember there was a complaint, a DOJ complaint, filed against Apple. You might remember that one was also very specific in terms of the behavior and the markets in which it alleged Apple was a monopoly.

And I think that's the case here with Google in terms of dividing it up into those different types of markets, specifically like the tools used by publishers, like new sites to host ad space, open ad space. Right, if you are the New York Times or you are wired or whatever, and you are, like, created using tools to be like here's where the ads go on our site and how that dictates this in some cases, what gets seen, or things like that. That's one market, the tool that advertisers use to buy that space. Right, if I am somebody making a product and I say I want to advertise in the New York Times or I want to advertise on Wired, you have to use Google's tools in order to buy that, those ad spaces and the software facilitating those transactions. That's largely, by the way, a summary not very nice summary given by the New York Times. And what they were talking about here specifically is the first one, where it's like, you know, the advertising space that's available and the software system that facilitates it. But not necessarily the case where I'm the person with the product looking to buy ads, which I think makes sense because you know you're looking to buy ads, which I think makes sense because you're looking to buy ads from Google. So you're kind of going into that already, being like, all right, I'm going to buy ads from Google. They don't have a monopoly overselling you the ad space that they're providing.

But the fact that they do control if you're a publisher, if you're a New York Times or you're Wired or you're whoever and you're looking at who do we want to buy from and partner with, to have ads on our site, there really wasn't an alternative, right, that's where the monopoly comes in. If you are publishing a website and you're a publisher and you're like we got to sell ads to make money and to generate revenue for our publication, you didn't really have a lot of alternatives. You had Google, and so that is a pretty good argument in many of those cases to say, look, there is a monopoly in that market because you really couldn't choose anything else. And so then Google gets to dictate things like ad rates and terms and where things go and the behavior of ads, all of those things. Let it exercise a lot of power in that area. And if the answer is, if you're a publisher and you're like we don't like the way that's being done, then your alternative is not go to another vendor, necessarily, but not have ads which is a problem.

0:09:21 - Mikah Sargent
Yeah, that's not an option for everyone, for anyone like that. That's part of the sort of agreement with the business is that you are going to tell people about it, and so, yeah, like there have been times where I have felt of two minds perhaps about the arguments being made, and then there are times where it's kind of like I get it, I understand where this is coming from, and this feels like one of those times. It's, I think, in part, though it's a bit of a culture shift, a legal cultural shift, if you will, where there was a time that big tech was getting off the ground in a big way and you did just kind of you got this and you got that and you built this and you and it's going oh, look at this cool stuff we're able to do, able to do. And that was enough to kind of obfuscate and not worry about being anti-competitive.

0:10:38 - Dan Moren
But now it seems like the focus is there and we're kind of reconciling as we're catching up. It was the nascent era, right, where it's like fast. I mean the same way that, like you know, when they first started building cars in the early 20th century, they probably weren't doing a lot of crash testing. But you know they are getting the whole industry running and then it's only probably later on. You're like maybe we should think about making these safer, right? Like maybe we should consider whether or not people might die when we crash our cars into one another, whether or not people might die when we crash our cars into one another.

And I'm not saying that's right, I'm just saying it is oftentimes. I think this is true overall, across the board, not just in technology, but oftentimes the regulations and the courts move slower right. Things can happen much faster because by nature, the law is a deliberative slow thing that needs to examine how things are and the way things are done and then concoct a strong argument based on the understanding of these things. And, by necessity, that is slower than just being one being like we're out here building stuff with code, or we're building cars and we're driving them on the streets right, it just doesn't. It takes a while for it to catch up in that, as well as so many other things yeah, exactly, including driving on the street without people in those cars.

0:11:53 - Mikah Sargent
Uh, which is a new thing that you know is is sort of we're still catching up to yeah exactly driving an ai system to do things that it shouldn't be doing. I don't know, it's all driving. Yeah, we of course will. As is always the case when it comes to the appeals process. This is going to be years long, likely, and involves so many different twists and turns, and you know, google is sort of juggling other anti-trust cases at the same time, so we'll have to see how this shakes.

0:12:34 - Dan Moren
That's happening soon too. Just as a footnote, the um I believe monday, the um, the previous case over search will now have a hearing about essentially what the Justice Department is going to do or ask for in terms of remedies in that case, which could include breaking up, you know, Google, which may play into this case as well. So we'll you know all those things are going to intertwine a lot over the next months and years and we'll see what happens.

0:13:01 - Mikah Sargent
All right, let us take a quick break so I can tell you about our first sponsor. Thank you, Dan Moren, for joining me for this week's episode. This episode is brought to you by ThreatLocker Ransomware. It's harming businesses worldwide through phishing emails, infected downloads, malicious websites and RDP exploits. You don't want to be the next victim. ThreatLocker's Zero Trust platform takes a proactive deny-by-default approach that blocks every unauthorized action, protecting you from both known and unknown threats. Trusted by global enterprises like JetBlue and Port of Vancouver, ThreatLocker shields you from zero-day exploits and supply chain attacks, while providing complete audit trails for compliance. ThreatLocker's innovative ring-fencing technology isolates critical applications from weaponization, stopping ransomware and limiting that lateral movement within your network. ThreatLocker works across all industries, supports Mac environments, provides 24-7 US-based support and enables comprehensive visibility and control. Mark Tolson, it director for the city of Champaign, Illinois, says quote ThreatLocker provides that extra key to block anomalies that nothing else can do. If bad actors got in and tried to execute something, I take comfort in knowing ThreatLocker will stop that. So stop worrying about cyber threats. Get unprecedented protection quickly, easily and cost effectively with ThreatLocker. You can visit threatlocker.com/twit to get a free 30-day trial and learn more about how ThreatLocker can help mitigate unknown threats and ensure compliance. That's threatlocker.com/twit, and we thank ThreatLocker for sponsoring this week's episode of Tech News Weekly.

Already back from the break, and now it's time to talk about social networks. What, yes? OpenAI, of course, is well known for its chat GPT chat bot is exploring what it might consider its next big leap, and it's not about language models. It's actually about building its own social network. So what exactly is driving this pivot? Is it a pivot, and what does it say about the future of tech's biggest players? Well, we know that, according to reports, sam Altman, head of OpenAI, is working on the idea of a social network, is in fact prototyping a social network that is built around AI. We hear that ChatGPT's image generation is a big part of this social network, and CEO Sam Altman is reaching out to folks and asking for feedback.

You know, this is sort of an invite only, kind of a deal where it's not even in beta or anything like that. It's like hey, friend, what do you think of this? You know it's interesting because at the same time this is going on, we have two other big names in similar spaces that are in headlines a lot. We have Elon Musk and we have Mark Zuckerberg, and each of them owns a social media network, and it just so happens that Sam Altman is also one of those names that is also in the headlines a lot and does not currently own a social media network. It would appear that OpenAI is kind of moving from just focusing on AI tech to diversify, but I wonder if there's something else going on here. I wanted to hear, though, first, dan, we talked about this a little bit yesterday on our show Clockwise and got a little bit of insight into what people were thinking about it, but with us it's been a little bit more roomy. You want to tell me what you're thinking about OpenAI's social media network?

0:17:16 - Dan Moren
I think it's worth taking a step back here and asking yourself why do they want to do this? Why do they want to do this? Why does OpenAI, a company that makes AI chatbots and other AI-based technology, want a social media network? And I'm going to give you a hint it's not because social media is a super profitable, super great business to be in in our year 2025. It's because it is still an AI company at the base and what it wants is data, and if you are going to spend your time on social media, that is a lot of free data.

I mean, I say this too we were just talking about Google and advertising, and this was always my argument to people who were like, yeah, Gmail is great, it's free. It's like, yeah, it is free. Think about why that is.

0:18:02 - Leo Laporte
Why is it?

0:18:03 - Dan Moren
free. Think about any of the products that Google gives away for free and why they are doing it. They are not altruistic. Same goes for OpenAI. Right, they are not in the business of hey, we're just here making cool stuff for the world, we don't really care about money. Sam Altman is a billionaire. He clearly cares about money. So when you sort of start that, take that as your starting point and saying the reason they want this is access to more data, then you know that may already, you know affect your decision as somebody who takes part in social media, to be like how interested are you in this thing?

I think that there's some real questions also of like starting a social media app in this day and age. Right, it is a very it's a very different market than it was a decade ago, in part because it is very established now. In a way, it was not a decade ago. Right, we have big players and it's very hard to come into a market like that and challenge those players, even if you are a company as sort of front of mind, as open AI.

You know the fact that you're going to unseat Instagram or TikTok or X is a lot. It's a lot to promise and a lot to aim for, and it's also a venture that, in and of itself, tends to be kind of a loss leader. Right, it tends to not be a big moneymaker because you're not charging people directly for it. Usually You're trying to make some money off the back end, whether it's by selling advertising or, in this case, by using your training data and kind of writing that off by saying, well, we're really improving our main product by using this. So I have a hard time thinking that this will actually be successful. It is possible, certainly.

I mean, like in terms of the way they've talked about hey, like maybe AI will help people post better things or something, and it's like I don't know if that's a good argument and you know people talk about Grok, which is you know X's AI thing that's been integrated with now and all of that, and this is a competitor to that in some ways. Yeah, I think this is very early days. Obviously I would be surprised if it actually even makes it ever to a real product, but it's certainly not something that I think a lot of people are like oh yeah, let me go sign up for that.

0:20:24 - Mikah Sargent
Yeah, so it's interesting. I have had on the show before I've talked about I think it was with Amanda Silberling on two occasions some social media or social networking apps that were AI driven One, where you create an AI character and that AI character posts, and then other people have their own AI avatars and those posts, and you're not creating, you're not creating a version of yourself, right? You're literally creating some other being, so you can you have some choices about. Oh, you know what's their personality like, what's this, what's that, and then you get to check on what they're doing and how. It's sort of Tamagotchi esque, but a little bit more more cognitive abilities, I guess.

And you. There was another one where you are a real person, but the site is all. Everybody else who you are following and who is following you is just AI. So you basically have your own bespoke social media network where you are posting, and it's real, but everybody else who's responding to you is AI, and in both of those cases there's, you see, like a little activity jump at the beginning, as people try it out, and then you see it kind of fall off and it's. I think the novelty of that wears off If, if we start, though, to see the introduction of these AI bots in a more, because there are currently AI bots where, if I were to post something and say, man, I wish I had a shirt of this, you're going to get five different bots responding to be like get it on Redbubble now, but not that kind of bot, but one that is more realistic, so to speak. If we start to have this happening more regularly, I'm kind of curious if people will get used to it and, in that way, kind of just expect that that's going to be the case.

Interestingly, though, I wanted to mention just yesterday there was a report about Meta blocking Apple intelligence on its various apps in yes, I did see this. Yeah, especially the writing tools. So we don't really know what's going on, other than if you tap on something and you try to use the Apple intelligence features, like the writing tools, the option is not available. It doesn't let users create and share Genmoji. Meta has also removed the ability to add keyboard stickers and Memoji to Instagram stories. Why that's the case, don't know, but I find it fascinating from the perspective of each of these companies. There's sort of a convergence going on right now, because at one point, we had the social media networks working to add AI, but now we're seeing the AI companies working to add social media and there's this sort of overlap and convergence that's taking place, even if we look at the goal behind it and, yeah, we look at the fact that so much of this is just about getting not just more data that's real overall, but also real-time data.

I think that's the big thing. What's the zeitgeist? What is the latest meme? If I want a relevant tool, it needs to have real-time information, and so you've got Meta working on an AI assistant app that's based on your feed and your interactions. You've got X merging with XAI.

Of course we know that it trains on real-time user content. We've got Google, who has its own virtual assistant that uses different things, including YouTube videos to train and whatever Android stuff is available. And we've got Apple, which just recently put out its own report about and, mind you, with certain privacy stuff in place, but still, if you agree to hand over your data via analytics, it trains on user email. It trains on usage of of Genmoji, and again, I want to be clear it's a little. It's a bit.

It's a bit of an oversummarization or oversimplification to say that it trains on user email, because it goes through all these different processes to make that obfuscated, but essentially it does, and so all of these companies, these big tech companies, are using their own analytics platforms to be able to train their versions of generative AI. So yeah, in some ways I wonder if OpenAI is not a little bit jealous that there's this user base that all of these companies have the implication that Grok integrating with X yeah, you're going well, I wish I could do that. Anthropic's probably going. I wish I had user data that exists outside of just the prompts that I'm giving. And then in many ways people are very sensitive to their prompts being used to train. So there's lots of like toggling that kind of thing off. So I'm not surprised to hear about this.

0:26:18 - Dan Moren
And the other thing that I'll mention is just we see the main big tech companies all trying to be that Omni app, that Omni service, yeahni service yeah, I mean, that's definitely the case for x, you know, which I think elon musk has said it's kind of his goal long term is to have it to be everything, and even recently he had this little financial shell game where he sold x to his xai company, which was a little confusing and weird, but, you know, again goes to the same idea. And the other thing that's worth noting about this is obviously a bunch of these AI companies are in litigation and argument over using data that they may not have permission to, whereas if you sign up for a social media network that's designed to feed data to an AI, you can bet that's going to be part of those terms that you don't read right as you scroll through and say oh, yeah, yeah, just sign me up.

I agree, you've probably just given consent to it to train all your posts and your photos and everything to be used to train AI. So that's the way it goes.

0:27:23 - Mikah Sargent
That's the way the AI cookie crumbles. All right, it is time to take another break before we come back with well more antitrust in just a moment. But first let me tell you about Melissa, who is bringing you this episode of Tech News Weekly. Melissa is the trusted data quality expert since 1985. With Melissa's launch of the Melissa Address Validation app in the Shopify App Store, shopify Plus, merchants now have access to a critical data quality tool directly from the address expert, and I'm so happy that that's the case, because I have had my bacon saved by this address validation in the past and I said you typed something in wrong and I said no, I did it. Oh, yes, I did, and so I got my product when I was expecting.

Key features of the Melissa Address Validation app include real-time address validation, so customers are immediately notified, like I was, if the address they enter contains errors or needs correction. Preventing shipping delays and reducing return rates. Global Reach the app validates addresses in more than 240 countries and territories, so it's standardized according to local postal regulations. Enhanced Address Correction even if Google Auto Suggestion is enabled, melissa's app can correct and standardize addresses, adding missing components such as postal codes, and ensuring compliance with local formatting rules and also CASS and SERP certification. The address engine is certified by USPS and Canada Post, offering reliable validation for North American addresses. And you can't forget the smart alerts. The app provides a warning on the thank you page if there is a potential issue with the shipping address, allowing customers to update their information before the order is processed.

Shopify Plus, merchants can easily install the Melissa Address Validation app to improve customer satisfaction and avoid the costs associated with returns and re-deliveries. With Melissa, you get secure encryption for all file transfers and an information security ecosystem built on the ISO 27001 framework, adherence to GDPR policies and SOC 2 compliance. Get started today with 1,000 records cleaned for free at melissa.com/twit. That's melissa.com/twit, and we thank Melissa for sponsoring this week's episode of Tech News Weekly.

All right, we are back from the break, joined by East Coast Bureau Chief of Six Colors, Dan Moren, and it is time to talk more antitrust.

0:30:02 - Dan Moren
Antitrust.

0:30:03 - Mikah Sargent
Antitrust.

0:30:05 - Dan Moren
Antitrust is my mom. Yeah, my mom's sister.

0:30:08 - Mikah Sargent
Antitrust is my mom yeah, my mom's sister. You know, mark Zuckerberg thought his political alliances and a lowball offer could help Meta escape a historic antitrust trial. Instead, the FTC called his bluff and now he's back in court testifying about the very acquisitions that built Meta's empire. The very acquisitions that built Meta's empire, of course, we know Meta. Formerly Facebook has been embroiled in a trial over the acquisitions of Instagram and WhatsApp, where the FTC says look, you've acquired two companies that make this space very difficult to thrive in. If you are not one of these companies. It was originally filed back in 2020. So it's been going on for a while. It was thrown out and then it was brought back into reality under Lena Kahn's leadership. Now here's the interest. Well, I guess a lot of this is interesting, but here's the the uh. Here are the details. According to uh, the wall street journal, um, meta offered to settle this case just $450 million, and maybe that sounds like a lot of money to you or to me, right? But here's the deal. The FTC said you are going to be paying 30 billion with a B dollars 30 billion dollars. And Meta said well, what if we just gave you four hundred fifty million? Meta said well, what if we just gave you 450 million? You know what if we just did that?

Then we heard that Zuckerberg was really cozying up to our current president to maybe let that $450 million be enough and not have to pay $30 billion, as things started to get a little rocky and it was becoming clear that the case was going to go through anyway. Meta said okay, look, we get it right. We've tried these different things. We've gone to the dinners. We've tried to give you less than half of $1 billion. So what if we just give you a billion? Is that enough?

No, the FTC said sorry, no, that's not enough, in fact. Fine, $30 billion, fine, okay, that's a lot. Right, we want at least $18 billion and we want firm restrictions in place. Well, that didn't go through either. So now they're in the trial and, as far as we can tell, the Trump administration has ultimately backed the FTC's position as it stands, and so all of that work and lobbying has not been has not led to to the outcome I think that Meta would want. Goes back to what we were talking about earlier with the previous antitrust conversation, where the culture was once one of you buy what you want, you get in trouble for it, and then you just settle, and then you move on, and it's shifting from acquire uh, get in trouble, settle to companies actually having to answer for the decisions and acquisitions that they're making. It's interesting, though, because Dan, correct me if I'm wrong but don't acquisitions have to be approved?

0:34:09 - Dan Moren
Yes, absolutely, that is definitely a. Thing.

0:34:13 - Mikah Sargent
Yeah, and so it was approved, and now they're getting the company's getting in trouble for it. Look, I'm not to be clear, again, not a lawyer, but it's absolutely an argument that there's a lot of anti-competitive behavior involved in this. And I'm not standing up for meta itself, but it is odd that you have to have an acquisition approved and then yet you can still get in trouble for it. An acquisition approved and then yet you can still get in trouble for it. It's like me saying mom, can I have a cookie? And mom saying yeah, and then later mom going, you ate that cookie, you're grounded what?

0:35:03 - Dan Moren
That's just not the agreement I need. That's true, but at the same time, it's also like can I have a cookie? And then what if you ate all the cookies? That's true.

0:35:08 - Mikah Sargent
You'd probably get in trouble for that.

0:35:10 - Dan Moren
I mean, it's a matter of degrees too right, I mean. I think you're right to a certain extent. Let's also remember that these deals are overseen by government departments that change over time, both in terms of the administration they represent as well as the people themselves who are at these organizations. So what one incarnation of the FTC might sign off on a decade prior? Ten years later, you might look at it and think, well, maybe that was a bad choice, and behavior is also part of it. Right, Like what they do with it after they've acquired it may influence that, because if they continue having, it's not just the act of buying it that is part of the entity competitive pattern right what you then do to discourage competition in other ways or to solidify your hold on it, or to lock down or remove things from instagram, right, that you might have had previously.

That it's like, well, we don't want other people to have access to our stuff now. So you know it is. It is a nuanced issue, I think, as we discussed previously with these kinds of court cases, but you're right that they got approval to do this at the time, so that does feel, you know, like well, there's some degree of blessing happening there. Why has this changed? And the answer is well, some of it's. The world has changed, right?

0:36:31 - Mikah Sargent
Yes, I think if I was a company that was thinking about making an acquisition, I would just do my best to avoid it if, at any time, I could end up getting in trouble for it. At the same time. Again, you are absolutely right in that it's At the time right. The approval for this specific thing may have made sense, and it was, you know, not as anti-competitive and, of course, the different people that were in charge, but, as you point out, the behavior that follows that to scoop up WhatsApp and Instagram and to to be as big as the Facebook part of Meta is is one aspect of that about the company's role in the information during the information that was spread during elections, right, and so there was that concern as well.

Plus, instagram has been in the limelight for for youth health, and Instagram has been in the limelight for, uh, algorithmic stuff. At algorithmic concerns, I mean, we've also seen Instagram working to compete more directly with Tik TOK. So there are all of these little bits and pieces that lead up to making it more likely that you're going to get in trouble. But the buy, lobby, settle aspect of the thing, the way things once were, has changed. It's not enough, it seems, to just lobby your way through and that I have to tell you that surprises me a little bit in the current administration, where one thought one could it kissed the ring, as it were, and that that would be enough to mean that, uh, zuck got out of having to go testify and and potentially spend a lot of money and that hasn't happened. Does that surprise you, or a?

0:39:11 - Dan Moren
little bit.

Yeah, I mean because I mean I don't want to delve too deeply into politics though everything is politics now, but certainly this administration has not seemed afraid of currying some quid pro quo.

What's interesting about this particular case is that, as you noted, it dates back to 2020, the very first incarnation of this.

There was an attempt from the FTC to sue Facebook in December 2020, which is the end of the first Trump administration.

That lawsuit was thrown out and then the FTC refiled a more detailed case in late 2021, which is the one that we're now seeing work its way through now seeing work its way through.

So there, and we know that there were indications at the time that part of what spurred that initial wave of, you know, attempts to sue Facebook was some degree of feeling aggrieved on the part of the both that administration and into, more generally, this idea of sort of censorship quote unquote, put that in quotes for reasons of right wing viewpoints and things like that, and that has certainly come up a lot in sort of congressional hearings and the likes. So, even with the currying of favor from Zuckerberg to the Trump administration, it does certainly seem as though maybe there's still some latent antagon being prosecuted under the previous administration that have just been abandoned or thrown out because the current administration, the fact that that gives them leverage over a company as big as Meta. I mean that is not a thing you can discount either that having them on trial very publicly and with big stakes does give them a lever to press in terms of getting what they want out of that company.

0:41:32 - Mikah Sargent
Big, well done stakes. Yum yum, yum yum with ketchup. All right, let us take a quick break before we come back with the final story of the week, again, this before we go to that break. This is always a matter of where will it end up, and so we can't. We can't know that yet, because there's always the process and the back and forth. We've had some great antitrust journalists on antitrust journalists on the show in the past, and I plan to have them on again in the future to talk about where things do end up.

All right, let's take that break so I can tell you about DeleteMe, who is bringing you this episode of Tech News Weekly.

If you've ever wondered how much of your personal data is on the internet for anyone to see, well, let me tell you it's probably more than you think. Your name, your contact info, your social security number, your home address, even information about your family members it's all being compiled by data brokers and sold online. Anyone on the web can buy your private details. This can lead to identity theft, phishing attempts, doxing and harassment, but you can protect your privacy with DeleteMe. You know, I, of course, am a person who exists publicly and being in some level of the public and out there, I do have to think about, you know, my online security. Well, that means that if I'm thinking about my online security, it's likely that my personal information is out there for people to find, or at least it would be. But that's why I recommend and use DeleteMe.

Deleteme is a subscription service that removes your personal info from hundreds of data brokers. You can sign up and provide DeleteMe with exactly what information you want deleted, and its experts take it from there. Deleteme sends your regular personalized privacy reports showing what info they found, where they found it and what they removed. And Deleteme isn't just a one-time service. Deleteme is always working for you, constantly monitoring and removing the personal information you don't want on the internet.

You know I went online to Deleteme, set things up and I immediately you say this is who I am. These are the things that you need to know about me. They get to work. Suddenly, I'm seeing the stuff that's been online and I'm going what in the world? Why is that out there, having been public and sort of having an online presence since 2011, 2012. I have, in the past, had individuals that were creepy reaching out to me and finding out information about me and I wish at the time I had had a service like this. And it's really nice now to have this service and say not only am I keeping my information offline as much as possible, but I'm also protecting my family, making sure that they are not linked to me as well.

To put it simply, delete Me does all the hard work of wiping your and your family's personal information from data broker websites. You can take control of your data and keep your private life private by signing up for DeleteMe at a special discount for our listeners Today. Get 20% off your DeleteMe plan when you go to joindeleteme.com/twit and use the promo code TWiT at checkout. The only way to get 20% off is to go to joindeleteme.com/twit and enter code twit at checkout. That's joindeleteme.com/twit code TWiT. We thank DeleteMe for sponsoring this week's episode of Tech News Weekly.

All right, we are back from the break, joined by Dan Moren, and we are now talking about a new trend over in AI land, so OpenAI-.

0:45:31 - Dan Moren
My least favorite theme park, my God.

0:45:34 - Mikah Sargent
Yes, none of the roller coasters actually connect.

0:45:37 - Dan Moren
Sometimes you just go and you're in that roller coaster and it just goes. There's no more rails, you just go.

0:45:42 - Mikah Sargent
It's so dangerous. Sometimes there are three different rails where it should only be one. It's a nightmare. So, over in OpenAI land, openai released announced its newest AI models 03 and 04 mini. These are reasoning models, meaning that to a certain extent they will take the request that you send them and then come up with a response. But look at the various responses that they could possibly give. I should. I keep saying they cause it's more than one, but like that, it could possibly give and then choose the best response from the possible responses. That is again, a simplification of the reasoning aspect of it. But along with being able to reason, it is also able to use all of the tools that OpenAI provides the searching, the web tools and the image integration.

And there's a sort of trend going on, and I'm going to keep calling it a trend because of something that Kyle Wiggers of TechCrunch has pointed out in Kyle's article that I'll talk about in a minute, but it essentially has resulted. I don't know if you've ever seen the wonderful and terrifying videos of people doing what's called geo-guessing, but essentially, for anyone who doesn't know, someone will take kind of an obscure photo of themselves or a scene and they'll send it to out of frame or is barely in frame, whatever it happens to be, to figure out where that person is. Well, people are taking obscure photos and uploading them to O3 and saying geo-guess this, and it is coming up with pretty good answers. In fact, in the cases that are shared, it is correct. In those cases, and this can be as simple as a photo of a let's see, it was a photo looking out a window out a window, and the person seems to be kind of sitting at a table and there's a coast nearby, and they correctly guessed that this was a person sitting in a Ritz Carlton in California, salt Creek Beach, and talk about all of the different information regarding that. Now some people might go wait, maybe the AI is just looking at the exif data. No, that is not the case. The folks are stripping out the exif data before they upload it, and it is able to use all of the different tools to do this.

Now here's the part where I want to make something clear. When we talk about this being a trend, it's not necessarily the fact that it's 03 is so much better at doing this, so much so that it is just people thought about doing it with this and then it caught on. People thought about doing it with this and then it caught on Because TechCrunch ran some tests using the current GPT-4.0, the model that has been around now for a while. That lets you be able to use photos and do the image reasoning capabilities or, excuse me, without image reasoning capabilities and it was still able to arrive at the answer in many of the cases, and sometimes it did it faster than O3 was able to do, which has a lot more going on. The reasoning models take longer because again it's looking at results every time and then kind of reasoning down what's the best answer from there.

So stalkers may have been able to use this kind of a thing for a while, and it just happened to catch on with this latest model because people are going oh, now it can reason with images. Maybe it's able to do something with that. I wanted to ask, dan have you heard about geo-guessing and kind of what's? What are your thoughts on this, as it starts to become apparent that we do have a sort of multi-tool in our pocket for so many different things that are sometimes a little bit scary?

0:50:32 - Dan Moren
Yeah, I'm familiar with geooGuessr. I haven't spent a lot of time with it, but I've checked it out in the past, this kind of thing. I think there's also some degree of this where this kind of thing felt like it used to be purely the province of humans, right, I was looking at an example of taking a picture. Somebody posted took a picture in a library of some books and it figured out where and that like seems bananas on the face of it. Like how could you just take a picture of some books and know what library it's in? But the answer was because it had the little labels on the side from the collection and they were labeled in such a way that it could decipher what it was. So when you think about things like that, you're like, oh well, that's kind of obvious and this used to be the province of people, right, people you know.

You think of this in like detective novels or you know movies, when people are trying to track down where something is, where is this picture taken, or what's that audio in the back of the phone call that lets me figure out where this person is calling from? It's no surprise then that there are tools that can figure this kind of thing out, but it is certainly something that should be sobering to people, right? Because I think it drives home the idea even further that I wouldn't say, like, don't share anything online, right, like that's an extremist point of view, but I think it reminds you to think very carefully about the things you share online and where and how you share them, because, as they point out in this TechCrunch article, sure, you might send something via Snapchat or Instagram story that's time limited and maybe even only sent to certain people, but that can be screenshotted and uploaded and shared in other ways, right? Or it can even you know what we used to call the analog hole, which is, you know, take a picture of it with another phone or another device. Right? Like, it's very hard, once that data is out there, to ensure that it can't be misused, and so it becomes then an issue of really being selective or willing to be okay with those consequences and realizing, like, look, I'm posting a picture of myself on vacation or something like that, and I know that it may be obvious from this, heck, I might share with you where this picture was taken, but being aware that certainly, obscurity is not necessarily enough to protect information that you want to protect, and in some ways I think that, like I said, in some ways I think this is good, because tools like this have been around, or less sophisticated methods of doing the same thing have been around, and so we have all gotten very accustomed, I think, to sharing a lot of information about ourselves publicly, and this is a reminder that these tools are only going to get better at sort of tracking down this stuff and that what you think might be a, if not private then at least a, you know, concealed piece of data may not end up being that private.

0:53:32 - Mikah Sargent
Yeah, and ultimately the fact that you know, for 20 or so bucks a month, people have access to this tool where before maybe you did have to hire a PI or something. I just yeah, you pull out your phone and the person you, the person that's being targeted is suddenly more available because they decided to share a photo of themselves on social media. Anthony Nielsen did a little test in the chat and posted a photo in Mendocino and said can you guess where in the world this is? It responded this looks like the town of Mendocino, california, on the Northern California coast. The image shows wooden buildings, coastal architecture and what appears to be the Mendocino volunteer fire department building on the right side in red and white. Um, and then said that the church-like building with a tower on the left is another recognizable landmark often seen in photos of Mendocino, and I can't quite tell. Does it what? What is on the fire uh department uh building? Anthony, what is the actual text on that? Because I'm curious, yeah okay.

And then he says I can't even read it, yeah um, and that's interesting too, because it's if it's pixelated, it's still able to kind of pull that apart. Right, well, that, and that is interesting too, because it's if it's pixelated it's still able to kind of pull that apart.

0:54:57 - Dan Moren
Right. Well, that, and that is the thing that is certainly possible in a lot of cases, you'll you know it can kind of reconstruct that it's much better at that than a person is going to be.

0:55:06 - Mikah Sargent
Absolutely, because that is also part of how it does. Image generation, like from a blurred blobs made into reality, of a sort Um yeah, fascinating, uh, but also terrifying, ooh. So you can do that, and just be mindful of the fact that it can do that. And um, yeah, uh, but this is the hard part, right? I don't really have advice to offer other than don't share your photos online, which just seems like, uh, I don't know, that's obvious, it's an obvious thing to say, but it's also kind of a ridiculous thing to have to say, um, but if that's the only way to avoid stuff like this, then that's the case. Little Bluetooth tracking beacons that can sort of piggyback off of the huge network of iPhones, ipads, macs that are all over the world, and so it creates, like this, instant tracking network to find an item that you've lost. Great.

However, upon the introduction of it, there was a lot of blowback from people who are advocates for those who are victims of stalking and domestic abuse. Within weeks had a response and updates to its technology to answer the concerns therein, and then continued to work to make that better and continued to consult with advocates to make the technology better, while also bearing in mind that, up to that point, there were many other tracking devices on the market that arguably had the same concern, paired with the fact that you could go to Amazon and type in cellular GPS and get dozens of results for little trackers that can do that. Sure, my point is, the company heard that this was an issue and then worked to correct it, and I don't see OpenAI, based on what we've seen thus far and its behavior thus far, being a company that will respond in such a way.

0:57:55 - Dan Moren
I don't think they have that concern. I don't. I think this is works as designed for them. Isn't this cool that it can do this? Yeah.

0:58:03 - Mikah Sargent
And that's upsetting, sure is, and that makes me feel a little powerless. But yeah, I don't know, I guess X at Sam Altman with concerns. Maybe that'll do something, I don't know what else will, because it's enabling some arguably dangerous things. So, folks, that is going to bring us to the end of this episode of Tech News Weekly. The show publishes every Thursday at twit.tv/tnw. That is where you can go to subscribe to the show in audio and video formats. I'll quickly remind you about Club Twit at twit.tv/clubtwit. Seven dollars a month, plus the return of our annual plan at twit.tv/clubtwit. With your club membership tweak free trial, by the way you get every single one of our shows ad free. You get access to the twit plus bonus feed that has extra content you won't find anywhere else and access to the members only discord server, a fun place to go to chat with your fellow club members and also those of us here at twit, and that warm fuzzy feeling in your heart knowing that you're helping support the work we do here on the network.

Just last night I was doing Mikah's Crafting Corner. We had a great time building the latest of the Lego succulents and it turned out great and we had some really fun cozy conversations. While we just sat back and listened to music and had a chill time, somebody said, wow, this is cozy AF. And I said that's exactly what I was hoping for. So if that sounds good to you, join the club. twit.tv/clubtwit. We love, love, love to have you. Dan Moren, thank you so much for being here with me this entire episode. If people want to keep up to date with what you're doing, where should they go to do so?

0:59:54 - Dan Moren
Thank you very much for having me, Mikah. Always a delight Love to talk to you. You can find me a bunch of different places. I host Clockwise over at relay.fm with Mikah every Wednesday. I do a number of other tech podcasts, including the Rebound, which you can find at reboundcast.com, and I am a contributor at sixcolors.com. You can find out all about this, everything I do, including my many books that I've written, which I if you're watching a video I display behind me in a bookshelf. They're kind of far away so they're hard to read, but you can find out all about those and everything else I do at dmorin.com.

1:00:29 - Mikah Sargent
Beautiful, so head there, people. You can also find me online at Mikah Sargent on many social media network or head to chihuahua.coffee that's C-H-I-H-U-A-H-U-Acoffee, where I've got links to the places I'm most active online. Be sure to check out Hands on Mac and iOS. Today, later today and, of course, every Sunday, you can catch Hands on Tech, where I take your tech questions and do my best to answer them. Thanks so much for tuning in. I appreciate each and every one of you and I'll catch you again soon for another episode of Tech News Weekly. We're going to have a guest host next week, so thank you all and goodbye.

1:01:07 - Leo Laporte
From Silicon Valley boardrooms to tomorrow's AI breakthroughs. If you need to keep up to date with tech, you need twit.tv. At Twit, we're tech experts who understand what's happening and can keep you in the loop, and we do it in a thoughtful, informed and fun way. Start your Sundays with this Week in Tech, a roundtable of tech journalists and people in the know with a rundown of the week's most important tech news. But that's not all. All week long, you can stay ahead of security threats with Steve Gibson and Security Now.

Keep up on all things Apple with MacBreak Weekly. Listen to the most informed Microsoft experts in the world with Paul Theriot and Richard Campbell on Windows Weekly. Our flagship shows feature tech's most respected voices, giving you the insight you need to understand and benefit from the changing world of technology. We give you analysis you won't find anywhere else, so you can make smarter business decisions and take advantage of the technologies transforming your world. Whether you're a CEO, it professional or simply passionate about technology, twit's network of shows gives you the edge you need in today's digital landscape. Don't miss a minute. Subscribe to twit.tv today.