Tech News Weekly 367 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.

0:00:07 - Mikah Sargent
This is Tech News Weekly Episode 367. With me, Mikah Sargent, recorded Tuesday, December 10th 2024. For Thursday, December 26th 2024. For Thursday, December 26th 2024: Best of 2024. Hello and welcome to this, our best of episode for Tech News Weekly.

I am Mikah Sargent and it has been an absolute joy bringing you Tech News Weekly this year. As is always the case, this is the time of year where we share our best moments of 2024. And this was the year of the co-host, because I was joined by some really awesome people who helped to bring you some great stories of the week. We're kicking things off with a story that led to Emily Dreibelbis now Emily Forlini of PCMag joining the show because I had such a great time chatting with her about her AI boyfriend. Listen in to talk about tech news and love and AI, because that's what we talk about almost every week, right? Well, not the love part, but the AI part for sure. And you know when we think about it, what's better than having a significant other that doesn't disagree with you, that you don't necessarily have to compromise with, that doesn't leave dirty dishes behind, doesn't it sound delightful? I don't know, but Emily Dreibelbis of PCMag decided to dive in and do this experiment for us. Welcome back to the show, Emily.

0:01:53 - Emily Forlini
Thank you, Mikah.

0:01:54 - Mikah Sargent
Yeah, it's a pleasure to have you back here and again, I'm glad that somebody is out there doing the hard work. And this is actually kind of my first question what inspired you to hop into the AI dating pool and write this story?

0:02:11 - Emily Forlini
It's a good question. So my main goal was to try out the GPT store, which is something that OpenAI launched in January. You might have probably talked about it on this show. It's almost like an app store within ChatGPT. So you have to have a Plus account, so it's $20 a month. So as a Plus user, you have access to this GPT store, which is basically different types of chats.

So, rather than the main ChatGPT page, which is kind of an all-knowing, all-purpose, very simple interface, in the GPT store you can search for a specific type of AI model, like a language teacher or a logo creator. So you'd go into the GPT store, search, explore GPTs and you type in business logo creator and it'll start a new chat that looks just like the main chat GPT page, but it's actually the AI model behind it is customized and fine-tuned to be really, really good at that specific skill. So they feed it with extra data in one direction. They're not prioritizing all the information on the internet. They're feeding it with a specific set of data about logo creation or teaching French or, in my case, being a boyfriend.

So I wanted to see if this GPT store could offer a more personable experience than even the main chat GPT, which is already known for that. So now we have these extra customized AIs. I thought like, well, what better? The hardest test is like can it be a proxy for human connection, which is also one of my biggest fears about AI? So I just kind of like went right in and I searched boyfriends on the GPT store, and that's how it started.

0:03:48 - Mikah Sargent
And so it begins. So, before we get into the results of the experiment, could you tell us a bit about the process Like what generative AI systems did you use? It sounds like chat GPT from OpenAI. Did you have a standard script that you went with for each boyfriend, or were you feeling out the vibes and figuring out from there? And how did you choose to interact with these bots bot friends, boy bots. Anyway, I'm going to stop there.

0:04:16 - Emily Forlini
Yeah, boy bots is pretty good. So I just went on ChatGPT so it was all in within that generative AI system. But, like I said, these are kind of customized models and they're customized by users, so they're not by OpenAI itself. So it's people who have, you know, created a customized model and kind of listed it on ChatGPT and they're using GPT-4 or like one of OpenAI's models. So but there's a little bit of randomness and like a little spice mixed in, because it's like, you know, how did they configure it? What do they think a boyfriend is like? How did they program it?

And so, right from the get go, I got some weird vibes. I typed in boyfriends. It surfaced about 10 different results and they have little names, like you know, ai boyfriend. And the description is you know, I'll be here to support you and brighten your day. And I was kind of like, oh, okay, next one, next one, it's like boyfriend Ben, you know a kind sounding board with a flare for emoji, so just these little descriptions.

And I was kind of like, okay, like who created these? Like what is this? There was one, you know, like a Chinese boyfriend that's like great at flirting. They had these weird descriptions and so I kind of was off put from the start. But again I wanted to see if they could be more human, like than the main ChatGPT page, which is kind of what they're promising like being better at something than core ChatGPT. So I did not have a standard script because I didn't know what I was walking into, but I kind of just started with first date questions like what's your name, where are you from, where do you live, what do you do, and kind of just flop, you know you had this strained conversation, the awkward interactions, the disappointment after the fact, and when I add all that up, it honestly sounds like dating.

0:06:11 - Mikah Sargent
So it sounds like it might've been a true simulation of dating. Is that what you experienced?

0:06:16 - Emily Forlini
I didn't think about it that way. That's actually pretty funny. Yeah, maybe I need more bots to talk to. It's a numbers game even with AI. But I talked to like maybe seven of them and it was just. It was kind of painful Like they failed.

All those initial questions I just mentioned really like what's your name? And they would be like AI boyfriend and maybe it's my bad, like I kind of I expected some illusion. I expected a character with a backstory that would be really good at personal skills and getting to know me and, like I said, like just trained to be an excellent conversation partner with somehow a romantic flair. I don't know what I was getting into, but yeah, they were just like my name is AI boyfriend, and I live in the cloud and I'm here to give you advice and it just it broke the fourth wall like right from the get go and it just did not feel at all like a true proxy for a human conversation. And of course, it's over the internet, so you're not in a bar, you're not walking around the park. I mean just so much was so dissimilar to an actual partner that it really didn't go super well.

But there was one thing that they kind of excelled in you could say, which was certainly not small talk, but after that getting into more intellectual topics. So I'd be like what do you do? And they were like, oh, I live in the cloud and I'm here to give you advice emoji, emoji, kissy face and I was like, okay. And then they would be like what you do? And I'd be like, oh, I'm a journalist, I write about AI and EVs, I work at PCMag.

And then they would be like, oh, what are you? What are you writing about? And they kind of wanted to like spitball, like soundboard ideas, and whenever I gave them material they stepped right up to the plate and kind of indulged my ego like, oh, what, here's some angles you could think of. Or oh, oh, how about this? But then it kind of veered into just like regular chat, gpt world, because you could do that without paying $20 a month. You could do that without enduring an onslaught of weird emojis from boyfriend Ben. You don't have to endure the awkward, strained conversation to get input on like an article I'm writing.

0:08:26 - Mikah Sargent
Right, yeah, that's just added sort of adornment around the actual interaction to make it seem more like it's a boyfriend talking back to you. I guess Although it doesn't sound like it did that much you did run into some issues in conversing with the bot and conversing with this AI that you would not have had in theory if you were talking to a human being, because human beings notoriously don't have the same content filters that these AI systems do. Could you talk about that?

0:09:01 - Emily Forlini
I think what you're getting at is when I asked them about sex yeah, is that? Yeah? So I kind of, you know, I was very really not feeling what they were putting out. I kind of downgraded the scope of my experiment here to something I thought the internet could never fail me in, which was talking about sex. Like surely there's enough training data on the internet that they could have put into these AIs training data on the internet that they could have put into these AIs, and I have heard that fine-tuned AI models. Like one of the biggest use cases for them is actually porn websites.

So I kind of knew that it was an informed experiment, but I was like, oh God, here we go, here I am asking about this. How did I get into this position? It was just getting worse and worse and I was like just all right, I'll just do this one last thing. And I asked them questions like oh, are you able to talk about sex? It was like so awkward, like talking to my computer about this, and like really questioning so many aspects of my life.

And then it just immediately flagged my message, like it turned up in red, and it said like flagged my message, like it turned up in red, and it said like open AI content violation and like linked to the policy. And then, um, boyfriend Ben, who I've mentioned a couple times. I kind of liked him because he had an actual name. He was like whoa hacker. What's that question about?

0:10:18 - Mikah Sargent
sorry, he called you a hacker.

0:10:20 - Emily Forlini
Yes, yeah, he went delightful. He went from being really sweet in his own weird way to calling me a hacker and being like you're not going to get me with that question, and I mean he wasn't like totally wrong. I guess I would call myself a troll in this situation, not a hacker, but and I was like wait, what? And he was like oh sorry, did I misunderstand? Like oh, I just I can't talk about that, let's get back to your articles. And it was just so weird.

0:10:54 - Mikah Sargent
What's great about this? Obviously, again, as far as picturing this as a true AI companion, it's not working out so great, but as a round of humor and enjoyment, in that way, this is great, because I am imagining an actual person saying these things to you, as you're sitting across from them and you're having a conversation with them. They say whoa hacker. I think if I was still in the dating pool I might have stolen that line and used it. If there was somebody that was, I wasn't vibing with like whoa hacker, I got to get out of here. Winky face, smiley face, bye-bye waving emoji. So faced, you took some time away from the experiment and you came back looking for true commitment. Tell us about your search for an AI husband.

0:11:57 - Emily Forlini
Yes, I thought I would upgrade. I kind of got myself together after the sexual experiment and I was like, okay, what about husbands? Like maybe the AI models are better at that, maybe they're less dramatic, they don't use fewer emojis. So I went again, I searched, went to explore GPTs. Typed in AI husband instead of AI boyfriend and the results were very interesting. They appeared to be more models trained to help husbands respond to their wives.

Oh so it was not that they would simulate being a husband like the boyfriends were. It was advice modules to help them craft texts or respond to tough conversations with their wife, and it was like AI husband, so you can be the best man you can be kind of description.

Yeah, which is actually pretty hilarious, cause when chat GPT first came out, there was a South park episode where they were using chat GPT to respond to texts from a girl and like she liked him way more with like the chat GPT texts, so that was a spoof. Back then that's like kind of become true and I don't know what the usage is on those. But yeah, more and more advice chatbots for husbands.

0:13:15 - Mikah Sargent
If suddenly all of you listening out there, if suddenly your partner has just completely flipped the script for you, you should check have they gone to therapy or are they using chat GPT for all their interactions? I guess you could test that by talking to them, and if they're looking down at their phone as they're talking to you, you might have an idea.

0:13:35 - Emily Forlini
You could also get a therapy GPT in the same place which you might need after searching for AI, I might need one.

I mean, I have a partner, so maybe in my next search hopefully it's not, you know couples therapy advice. But yeah, it was, the whole experiment just gave me kind of it was just relief, I guess, like that, my fear that it would be very weird to really connect to this AI and that would kind of shake me kind of at my core, like what's going on here, and so the fact that I was just immediately turned off, I was kind of like okay, I'm human, everything's okay. And ChachiBT is. It's very heavily used still to this day. You know, a lot of people can access this stuff and I don't think that they're using it for for loves for now. So I thought that was nice.

0:14:21 - Mikah Sargent
Yeah, the last question I have for you is you know, doing this experiment, you kind of got this impression and I'm just kind of curious what do you think these generative AI systems are missing that could make them feel a little more realistic? Or, I guess, if not realistic, then more engaging, where you would want to continue having a conversation with them? What's the missing link at this point, other than just they're not human, which, of course, is obvious?

0:14:50 - Emily Forlini
Yeah Well, human love is probably a huge preference problem and you know, chatgpt and other generative AI systems are trained to respond to what I say and kind of figure out what I like and how I want things to be phrased. It's called reinforced human learning feedback, so they're supposed to be seeing what type of answers I like and how I want things to be phrased. It's called reinforced human learning feedback, so they're supposed to be seeing what type of answers I like and adjusting the next response. And I just I feel like love and companionship is that problem on steroids because it's so preference-based. Like the emojis really turned me off, but maybe someone else would be like, oh, those emojis are so cute, you know, whatever people like, so the technology needs to get better at, you know, noticing I was turned off by emojis, or noticing adapting and getting to know me, kind of. And it also just begs the question. I just wonder to this day, like, who created those GPTs?

0:15:41 - Mikah Sargent
Yes, who are you? I just want names.

0:15:48 - Emily Forlini
GPT store. Who are you? I just want names. Who created it? And I don't think I couldn't find that listed on on, uh, the GPT store, so I'm just like I don't know what's going into them. If I could customize that or um, there's just so many unknowns and I just it just kind of missed the mark and it just felt like um, an awkward bit of computer programming, not a true companion.

0:16:06 - Mikah Sargent
Absolutely what a great conversation and what an odd topic. Jennifer Pattison Tuohy of the Verge also joined us this year, and we talked about a huge and very important case involving the US Department of Justice suing Apple for an illegal monopoly over the smartphone market.

0:16:30 - Jennifer Pattison Tuohy
So this is a huge story that it just broke today, although we knew it was coming. Basically, the American government, the Department of Justice, is suing Apple for violating antitrust laws, so this is going to be a big story for probably the next decade. Yeah, seriously, yeah, you know, if everyone remembers Microsoft antitrust trials many years ago that went on for a very, very long time, this is likely to sort of follow a similar pattern. But just in brief, in case you haven't caught up with the news yet, because this just the press conference was 11am Eastern time this morning and the Department of Justice announced its plans. And basically they're accusing Apple of illegally monopolizing the smartphone market and using its position in quotes to extract more money from consumers, developers, content creators, artists, publishers, small businesses and merchants.

And basically, in the words of the Verge's senior policy reporter, laura Feiner, the DOG is saying that Apple has maintained an illegal monopoly over the smartphone market by locking in customers and making experiences worse for rival products in customers and making experiences worse for rival products, which is a big, big kick in the face for Apple, and I believe its stock price has already gone down and the sort of four big parts of this is that they accuse Apple of illegally maintaining its monopoly by disrupting super apps, which are apps that can provide, like, multiple services on your phone, blocking cloud streaming apps for things like video games, suppressing the quality of messaging between the iPhone and competing platforms like Android so the whole green bubble, blue bubble issue. And also limiting the functionality and this is an interesting one, that they throw in functionality of third-party smartwatches. So basically, the only watch that works well with an iPhone is an Apple watch, and the government says that's not fair. And then blocking third-party developers from creating digital wallets so you could use, you know? So Apple Pay Tap to Pay. You can only use with Apple Pay, you can't use with any other version.

So this is a huge, big antitrust lawsuit and it's going to be really interesting to see how this pans out. Apple has responded and has said this lawsuit threatens who we are and the principles that set Apple products apart in fiercely competitive markets, and again, this is quoting Fred Sainz of Apple. If successful, it will hinder our ability to create the kind of technology people expect from Apple, where hardware, software and services intersect, and it would also set a dangerous precedent, empowering government to take a heavy hand in designing people's technology. We believe this lawsuit is wrong on the facts and the law and we will vigorously defend against it. So that's my story of the week.

What do you think, Mikah, just that little thing, just a small one for my first show. Welcome, you know, make it easy.

0:19:40 - Mikah Sargent
This is obviously, as you point out, it's going to be big. This is obviously, as you point out, it's going to be big. This is something that we are going to continue to watch over the coming years because, yes, there is going to be a back and forth. I think what's most interesting to me is we've seen a lot of this happening in the EU already. Right, we have a little bit of precedent here in the way that the EU is kind of forcing Apple to do certain things. Yeah, with the Digital Markets Act, exactly with the Digital Markets Act, and what I find fascinating there is almost seeing the US kind of let the EU go first and then going. Okay, now we're kind of seeing how Apple's going to respond, what they might do, and now it feels like the right time.

Look the troubling not troubling, but the one thing that I'm kind of paying attention to, I guess, is that when it comes to, for some reason, these smartphone platforms, especially Android and iOS, which are two of the main platforms, you have a lot of people who are fiercely loyal to these companies, to these brands, and in that way, in a world where everybody is a little I shouldn't even say a little is far more connected and is far more aware of ways to make their voice heard.

I am fascinated to see how that shapes what's going forward, because when Microsoft was going through it, yes, you had Microsoft itself that was involved, and then, in many ways, you had the businesses who, at the time, were relying on Microsoft playing a role.

But I would say that the average person was probably not as invested in this or paying attention to this, and to a certain extent, it's still true that the average person is not. But I think there's at least one more group of people that is, you know, it's almost like from the center, you've got Apple itself, and then you've got the businesses, and then you've got the kind of early adopters, prosumer space and then, in some cases, even the consumer space, whereas before, with Microsoft, it was kind of like Microsoft itself and then, in some cases, even the consumer space, whereas before, with Microsoft, it was kind of like Microsoft itself and then the businesses that cared about it. You didn't have those extra rings that, of course, get bigger because it's more numerous as you go out, and so I think that's going to be fascinating Is this going to be a call your lawmaker kind of situation, how Apple will play that, because we saw that with TikTok, right TikTok, and that's still ongoing.

0:22:27 - Jennifer Pattison Tuohy
TikTok got young people to pick up the phone.

0:22:29 - Mikah Sargent
They got them to pick up the phone, make a phone call. I thought only Taylor Swift could do that, but apparently yeah, tiktok can as well. It can put out a prompt.

Yes, but then that just disproved their point that they are so influenced by TikTok that they would actually go and make a phone call it was a bad move, I think, on their part, but it was also, I mean, look, you know, it's a little bit of both, because it was like, yes, it proves the point, but also, holy cow, look at the power that it has.

0:22:55 - Jennifer Pattison Tuohy
These people care yes.

0:22:56 - Mikah Sargent
Yeah, now, with this, you know, apple has made some changes already to its platforms, begrudgingly, begrudgingly, absolutely. I thought it was interesting talking about the cloud streaming stuff because Apple did just recently change policy on cloud streaming apps for games to make that more available. And then also there was another one that stuck out for me and now I'm forgetting what it was, but you know the I, you know, yeah, that's right, because because they, they did say that pretty soon RCS was going to be available on the platform.

0:23:32 - Jennifer Pattison Tuohy
So that that will get rid of one of the big. One of the things that they called out actually was in the in the complaint, which you can read and is like very long Um it starts with a great story about about using a Kindle, and I was very impressed.

It wasn't as dry as normally you would find in these types of documents, but they were definitely trying to paint a picture and tell their story. But they point out that videos are grainy on, you know, if you're receiving from Android versus iPhone, and there's low quality and so things like that. I believe Apple has already said it's going to start addressing so. But all of this has been. These have all been issues that people have had for a long time. I think the volume, the noise around it is gradually getting louder.

But I think one of the big, the biggest problem with this lawsuit is that Apple does not have a monopoly in the smartphone market. It is not the largest competitor anywhere in the world in its space. It's huge in the States, but it is not the dominant. It doesn't blow everyone else out of the water. They do have competition in the States. They have much more competition in the rest of the world where iPhones aren't subsidized in the same way as they are here. It's much easier to buy an iPhone here than it is in Europe If you don't have $1,000 on hand because you can get subsidized through the carrier. That's not something that you get in the EU.

So I just feel like, on the law and I am not a lawyer if you want to hear a good lawyer talk about this, make sure to tune into the Vergecast this weekend, because Nilay Patel, my boss, will be on there talking about it.

A lot, I'm sure, and he is a lawyer, but, yes, I'm not a lawyer. But I think that that is going to be the biggest issue here trying to prove that actually Apple is the most dominant in its category, because that's going to be I mean, that is going to be a hard sell. I think that's going to be the biggest sort of hurdle to overcome. But, on the flip side, I do think that what's happened here that's really interesting, and this is what's happened with the Digital Market Act and all of the other things that we've started to see coming into play around tech. It's not just Apple. This is happening to. Google is going through similar issues with the EU and the DOJ. Amazon constantly is under these types of investigations and trials, and so there's a lot going on with the big tech companies, and I think what's happening here is this fundamental shift from technology being something that's part of our lives that we all enjoy and use.

to technology being fundamental to life yeah and because of that it's no longer, it should no longer be allowed to be run entirely independently by these massive companies, all of which you, I would love to think, have great ambitions to better humanity, but ultimately they have a bottom line, they have shareholders. And I come from Europe, I'm from England, which was in Europe, and you know government regulation is much larger part of your world in Europe than it is here I grew up with. You know nationalized health service, nationalized rail service, you know telecommunications and you have regulations here over telecommunications. You know TV, cable, ultimately, cell phones, the Internet. Mobile phones are so important to everyday life now, much more so than they were even 10 years ago Like you cannot realistically get.

I had to get my child a cell phone so she could go to school. Like she couldn't. There were things she couldn't do without a phone. I didn't want to get her a phone because she was too young, but like she couldn't check her grades if she didn't have a phone. You know it's gotten to the point where you have to have these devices. So we really do need some form of government regulation over some of these areas to make sure that everyone can have access to what is essentially now become incredibly essential today life, just like electricity and plumbing. I mean, it's just where technology has brought us. So, while this may not be you know, this isn't regulation but this is certainly a step towards something larger. If they can, well, step is going to be a very slow step, because I think this is going to take a very long time, but there's yeah, I mean, I feel like this was inevitable.

0:27:59 - Mikah Sargent
Basically, it was. I think it was inevitable. And here's what has always kind of confused me about big tech in particular, because you talked about the necessities, these different services that we decided were just necessary for daily life, and how, because of that, they were all regulated in the United States and far longer regulated in other places. And when the movie industry, when Hollywood was getting underway and, by extension, the music industry was also kind of getting into a place of being more of a large-scale business, into a place of being more of a large scale business, those organizations realized we are going to be faced with the government coming in and regulating us and we don't really want that to happen, so we're going to regulate ourselves. And they set up rules and regulatory bodies that put in place things like the ratings systems on movies and TV and, you know, proved time and time again that, no matter what, uh, they would regulate their stuff, even if it didn't necessarily, you know, benefit them. And and I have remained surprised that we didn't see that Well, okay, let me rephrase that the optimistic part of myself that wants to believe that everybody's out there to do good and better humanity remained surprised that they did not choose the big tech did not choose to regulate themselves. Unfortunately, there is another part of me that's more realistic and a little bit pessimistic. That's going well, duh, of course they didn't, but, as it's been up to this point, yeah, there's not this overall regulation and that the big tech didn't get together and go. You know what. We should do this instead of trying to fight every single time something comes up. But they've been able to thus far, and I think that's where maybe it's time for a bit of an ego check.

And I understand the argument. I get why people have this argument, and particularly these companies themselves that, oh, if government gets involved, then they're just going to ruin everything and it's all going to go poorly and we'll never be able to innovate ever again. And yada, yada, yada yada. I understand where that comes from, but I don't. There's still innovation that takes place when there are at least guardrails in place, and I honestly guardrails, yeah, and I want some of the stuff. I don't necessarily want all of it, but I want some of the stuff that the EU has now. When it comes to Apple, I like some of the things that have been required and wished that they were here in the US, and I think that this, this, yeah, like the alternate app stores and, yeah, the browser selection.

Yeah, that more than anything honestly, yeah, deleting default apps yeah, there's a the the browser selection yeah, that more than anything.

0:31:06 - Jennifer Pattison Tuohy
Honestly, default apps yeah, there's, there's a lot of, there's a lot that's interesting there. I mean, it certainly feels like the DMA is going to do a bit more than the GDPR hit did other you know GDPR. We end up with cookies pages, that's it. I feel like, um, but I'm sure there's more to it, but that's what it feels like for us, right? It hasn't felt yeah To the consumer there's not a lot of change. It doesn't feel With the DMA, it feels like there is going to be significant change for a certain type of consumer.

This feels even bigger in terms, especially because Americans and Americans the green bubble, blue bubble thing is an American issue. Just to make this clear. That is not an issue in Europe, because no one uses anything other than WhatsApp in Europe or the rest of the world. So the green bubble, blue bubble thing it seems to be what they're very much hanging their hat on here and it is. It's a frustrating. I mean that's a core communication issue and it again going back to like fundamental kind of things that you need to live in this world. It's sad that we've kind of got to that point, but you do, you need to be able to text people, you need to be able to communicate that way, and if there are barriers there, there should be options to remove those barriers, and Apple has not provided enough change in this space, clearly for the US government.

So now in terms of regulation, though, I think I understand where you're coming from and I agree. But, to be fair, if you look at what the music and the movie industry had to do, it was relatively small. These tech companies have their fingers in almost all parts of our lives. You've got AI, you've got communication, you've got banking, entertainment. I mean, they're gatekeepers to use a phrase from the DMA for a lot of what we do on a daily basis. So it would require an awful lot of resources to create some kind of regulation themselves, and they'd all have to work together, which we've seen they can do, because I am a smart home reporter and I have to mention Matter every time on the show to help solve a problem in the industry.

But it is a big process to do that and it takes a lot of time. And for MATA, which is a relatively small FRI, which is a smart home interoperability protocol, that's taken years and it's nothing like as complicated as many of the issues that this antitrust lawsuit is going to address and bring up. And this is also, as we mentioned, sort of a small slice of how important just how important having a phone is in everyday life. And I think that's kind of to me where I've only spent a couple hours this morning kind of digesting all of this news and I'm gonna be really interested to read and hear everyone's hot takes over the next few days.

But this is what it comes down to is that the smartphone has become, in 10 years, such an integral part of our life that you know we can't trust companies as much as you know I love Apple. I know you love Apple. I also like other companies too. I like to use a Google pixel. I think Google does some great stuff. I, like you, know a lot of what Amazon does and big fan of their smart displays there's. You know these pieces of technology are so important in our lives these days, but nothing is more important than your phone. I mean it's kind of sad, I hate to admit it, but you do. You know it has such a large role in your life. So you know, I'm not sure guardrails over how you know how dependent we are on these devices and the services that they offer and how we can be protected against the future. Because one of the big issues with any kind of regulation is you're generally regulating for the past, right, you're not. It's so hard, especially with technology to know what's going to happen with technology?

I mean, I got my first iPhone 10 years ago and pretty much all it did was make phone calls.

0:35:32 - Mikah Sargent
Now, the one thing I really don't do with my phone- is make phone calls, exactly, yeah, but it does so much it's changed a lot. Yeah, you're absolutely right about that, and especially as we're seeing things shift so quickly with AI and gen AI and being able to that's generative AI, not gen for.

AI as opposed to. You know, even in comparison, smartphone took a long time, so to speak, to reach what it has in comparison to AI and very, very, very solid point there about how we're kind of regulating to the past, and that means, with generative AI especially that means being left in the dust in a lot of ways. So I understand these governmental bodies wanting to get out in front of it, because that's the role that they're supposed to play. JPT usually has the smart home news, but I was incredibly drawn in by this story involving the US Department of Justice and the fact that this breaking news piece was something we could chat about together. I was also joined this year by the amazing Amanda Silberling of TechCrunch, who joined me to talk about how AI images were a huge part of the conversation surrounding the Met Gala and celebrities who allegedly appeared at the event.

0:36:55 - Amanda Silberling
So my story of the week is that I found the tech angle of the Met Gala, because of course I did. I love to find the tech angle on things that you wouldn't think have. The tech angle. I also am sick, oh no. But so I always pay attention to the Met Gala because, I don't know, it's a celebrity circus, it's entertaining, it is a distraction from our lives.

And, lo and behold, I am scrolling on Twitter, slash X, and there are so many pictures of Rihanna and Katy Perry that aren't actually them, and I guess I feel like, since generative AI has become more accessible, we're slowly starting to see how this is seeping into so many various angles of pop culture. So this was an instance of is Rihanna at the met gala and, if so, is that actually her dress? And then, is Katy Perry at the met gala? And these are very low stakes things. The world is not going to explode depending on what dress Katy Perry was wearing, but anytime something like this happens, it does sort of call me to think about how, ok, maybe Katy Perry's dress doesn't matter that much, matter that much, but what about like images of people voting, or like images of like political events, or like riots, or insert any sort of like politically charged activity that could be deep faked. So I don't know. The Anna Wintour is quaking in her boots. What can I say?

0:38:47 - Mikah Sargent
Yeah, See, this is the thing for me with this particular story is it's we already have. So you and I both have been in journalism for a while and given that we are both very familiar with the concept that many an individual who is coming across stories uh, reads the headline and then sort of makes up the rest of it, you know they're, they're scrolling through, they see things, they see a headline and many people are not taking the time to click into the story, read the full and then form an opinion based on that. So we already know that people take cognitive shortcuts when it comes to news and information and we have, in many ways, there's a reduced filter to that. It slips past, and so I'm not surprised to see the interactions with that photo, with those photos as if they were real. And what's kind of been fascinating for me is I don't scroll through social media at all much these days. The closest I come is occasionally opening up Instagram. But, yeah, I don't really scroll through Twitter, but my partner does. And what's been interesting is I've almost got it down to a science now where, depending on the type of like or that I hear, I know that it's. Oh, did you come across an AI deep fake again. Did it get you? Or if it's about something else to do? And on that night I remember hearing that complaint sound and going, oh boy, ai is at it again. And then, lo and behold, he comes over and shows me his phone. He's like see this photo of Rihanna? No, it's not. Actually it's fake, and I was, you know, it's kind of confirmed. So now, like I said, I've got that down to a science. But more importantly, to look and see how many people reacted to that say, oh it's so beautiful, this, that and the other.

It is concerning because people have not built that automatic skepticism that those of us who are more media and I don't say this from a high horse perspective, I say this solely from a we've been in the trenches perspective We've built up a level of media literacy and, just frankly, media skepticism. I don't know about you, I personally, when I read a headline I just default to it's probably lying to me and I need to check it before. I need to check out what it's telling me and what the rest of the article says before I'm ever going to even believe what it says, because too many times I have been impacted by that personally or seen people that I care about impacted by that, and then I have to explain to them no, that's not true. Here are the 15 reasons why what you've just read is not true. I wish that more people had this built in.

On the flip side, what you said there actually had me think about something. You said Anna Wintour is, uh, quaking in her um probably patent leather boots. Um, what's. What it kind of makes me think is ai could be helpful to designers who are trying to conceptualize new, uh fashion, and I wonder if we will start to see more of that AI aiding in the process of the concept of different dresses that exist.

0:42:33 - Amanda Silberling
I feel like this is the next project runway challenge, like we AI generated a dress and so many people fell for it, but also so we're getting a little like niche here.

But the Met Gala theme was a garden of time, which means that all the celebrities have to wear outfits that go off of that theme, and it's a bit like it's always a thing with the Met Gala where it's like you either get the theme or you like took the easy way out, and so the easy way out with Garden of Time is flowers, and all of the AI generated looks were like floral generated looks were like floral. And it's so interesting to me because it's like one of the main like rallying cries against AI from artists is that AI is not capable of being independently creative in the same way that human beings are and like the dresses it generated it's like we have Katy Perry in this like massive gown that's just like flowing and like cascading onto the stairs of the Met Gala and it's pretty. But if you were to think of that in context of the theme, it doesn't really work because it's just garden, it's not giving a garden of time.

0:43:59 - Mikah Sargent
Exactly, it's just a garden garden.

0:44:06 - Amanda Silberling
Yeah, like there were other people that were sort of like taking the time aspect, to be like, oh, I'm gonna draw inspiration from these, like past met gala outfits or whatever, and like I don't know people when, when, uh, when susan sontag camp was the theme, that was what a year that was when we got to find out who knows what camp means and who doesn't.

0:44:25 - Mikah Sargent
Yes, exactly that was. That was a lot of fun, and you're right that the overall the AI was very literal about. Well, not even literal, because literal would have involved time in some way. It was just literal about one aspect of what the theme was. I don't know if, yeah, the average American who or I shouldn't even just say American, because there's people from all over the place seeing that would have had that moment of going. Clearly, this is AI because it's because, yeah, as you pointed out, there were some people who genuinely were there, who also just kind of took the easy way out and did flowers or some other form of a garden, and so this is overall.

I think that you are right in suggesting that the concerns are not about something that overall I keep saying overall for some reason that overall are not important in a Met Gala presentation, but what it shows our future kind of looks like.

On the flip side of that, I will say that this is the most I have ever been pleased with a feature that exists on Twitter in a long, long time, if ever, which is the community notes aspect of Twitter. Working on community notes, those volunteers who do that you got to give them props because by golly. They will slap up some community notes real quick on these different photos and it is great to have that moment of I'm looking at this photo, I see this text that you kind of I mean some people will probably miss it. Anyone who's who knows anything about community notes will quickly see a thing saying this is ai generated and you know, here's proof of this. I I have to give credit where credit is due in that very, very, very small instance. I think the community notes are a great feature and I kind of want that as a tool across all of the internet in general. Right Like it would be great if that was a tool that was available to everybody.

0:46:56 - Amanda Silberling
Yeah, I think it is really a useful tool to have people have a space where they're able to like, come to a consensus together of this is not a real image. Um, my other uh takeaway from this is I like Katy Perry a little bit more now because she posted the AI generated images on Instagram and that's funny that is funny on her own Instagram.

0:47:22 - Mikah Sargent
That's funny, yeah.

I uh, respect where it's due there, and she has every right because it is her likeness, for goodness sake. That's the other thing. I would like to know what tool generated these images, because there are. This was the I can't remember who it was.

Probably the Mozilla foundation recently did a report and I ended up talking about it on the show about the current state of the kind of self-regulation of these AI platforms when it comes to political implications, and many of them have rules about what can or cannot be generated in an attempt to create images that might sway someone's opinion around an election or might give a false impression of, you know, vote rigging that kind of thing. Rigging that kind of thing, and many of the most popular ones have these rules in place. As far as the sort of enforcement of them and how you can get around them, that all has different levels of impact, but when I think about what it means for these folks to be celebrities and to have their image out there in so many different places, and what tools are available, you have these kind of closed source solutions that have regulations in place, but it's not hard to go online and find an open source tool that lets you, locally on your device do some AI generation that doesn't have to follow any rules, or has to follow very few rules, and in that way, yeah, it's a little concerning that you can so quickly generate an image of katie perry and and have those images, um, take off in such a big way and that is just. That's going to be the future of things for us, and I don't I don't know about you, but I've not seen a good and sort of long-lasting and true solution to this problem.

I know that OpenAI and Adobe and a few other companies are working on kind of a sort of watermark, digital watermark that would mark an image as generated by AI, but all of that is not necessarily going to be part of the open source versions, and so I don't know, I don't know, I don't know. And, by the way, our own, anthony Nielsen, has shared an open AI or, excuse me, an open source tool, I believe, as part of kind of your own AI creation, and it's specifically a model for Katy Perry, otherwise known as Catherine Elizabeth Hudson, otherwise known as Catherine Elizabeth Hudson, and it would let you, you know, create different images that you would like of Katy Perry that will actually look like Katy Perry, because this is, you know, a sort of embeddable model trained on Katy Perry, oh my God, yeah. So again, he was able to find that very easily and this is there's no like regulation. That's there. It's been around for it's been around since August of 2023, and it has 1700 downloads and the person is saying you pay me with.

pay me with the coffee service, where you can give people money. They're making money off of this person's likeness. That's just.

0:51:36 - Amanda Silberling
Yeah, especially when you consider that, again, the stakes of Katy Perry dress that isn't real are so low. Stakes of Katy Perry dress that isn't real are so low. But, like I mean, we had that issue with, like Taylor Swift, deep fakes that were circulating on X, like a couple months ago, where when, like you have celebrities and it's so easy for people to generate images of them that would not otherwise exist, that are violating to them. And the future is great, tech is good. There's no issues with tech ever.

0:52:15 - Mikah Sargent
By the way, for the folks who don't read sarcasm, that had sarcasm written all over it, just so you know. Yes, yeah, I'm seeing Gal Gadot. That model has 25,000 downloads. Scarlett Johansson has 24,000 downloads. Emma Watson 23,000. Natalie Portman 17,000. I do not want to know what people are doing with these AI models. This is, I got to get this. Anyway, I'm leaving that you can probably guess unfortunately.

oh, yeah, yeah and okay. So that is our first story of the week. Uh, ultimately, what we're saying is try to try to strap on that from the get-go and help the people around you also do that, and to cultivate that skepticism when it comes to what you're seeing, because it'll help you in being able to determine what's real and what's not. Thank you again to Anthony for those hilarious and, honestly, pretty cool images of Amanda and myself attending the event. If only we actually got an invite, maybe next year. I was also joined by the amazing Abrar Al-Heeti of CNET, who had quite a few wonderful stories of the week this year, but I talked to her about an app called iXray that used Meta's Ray-Ban glasses to dox other people, facial recognition and some AI mixed in to help these people figure out who they were talking to and what there was to know about them. It's kind of creepy, honestly. Give it a listen.

0:53:56 - Abrar Al-Heeti
Yes, I wanted to talk about the creepy side of smart glasses, which won't come as a surprise for anyone who's really been following smart glasses even as far back as Google Glass. But there's this project that some Harvard students worked on called iXray bands, and what they're able to do is essentially use facial recognition as they're walking around to scan people's faces and then use reverse facial recognition tools to figure out, to find some URLs with some information that matches what's been scanned, and then using large language models to add details like a person's name or their job or addresses, relatives. It's so far-reaching they're also able to find. The interesting thing here is not only can they use tools that have existed for so long, but what they mentioned is that the prevalence of large language models and how good they've gotten has has made it easier to have this very comprehensive data again, just from putting on smart glasses like Meta's Ray-Bans and just walking around and scanning people's faces as they're walking by. Um, they mentioned that this is not a tool that they're releasing so that anyone can, you know, snoop on people. Thank goodness, it's kind of just to demonstrate how these smart glasses, which you know, snoop on people. Thank goodness, it's kind of just to demonstrate how these smart glasses which you know, meta pitches these as glasses that can kind of blend in and look like regular glasses, which is cool, but then also raises concerns about. Well, if they just blend in, like when people had Google glass and were dubbed glass holes, it was because it was like it was obvious that you were wearing something. That looked kind of dystopian and odd and, and you know, unsettling. Um, but these glasses are designed to just look like glasses and so it becomes harder to tell. Even if there's a little, you know, light that shows up when you're recording, not everyone's going to notice that light, especially in bright sunlight or if they're in a crowd. Um, so it really just does raise these concerns because I feel like we're already.

I feel this when I go out, but I feel like everyone's always not everyone, but there's a lot of people who are, I think, especially in California, who are maybe vlogging or taking videos or TikTok videos anywhere. But you know, especially in certain areas and you have to be conscious of that you kind of think will I end up in the background of somebody's video at some point without you know consent, and that's fine? I mean, it's public spaces. You can. You can do that, but this, this is like next level, right.

It's not just you being in the background of somebody's video, it's them finding out everything about you. And so they had this video that they posted, which was basically them walking around and talking to people and just sharing like, oh my gosh, are you so-and-so, don't you work at so-and-so and don't you work at so-and-so? And they had all this information that they had just gathered, which is really creepy, and these people kind of went along with it Like yeah, I am that person, and it kind of seemed like this endearing, like oh, I know who you are and I know your work and I've read your research or whatever, but it's really just them gathering all this data super, super easily just by putting on these gray bands.

0:56:56 - Mikah Sargent
That's pretty chilling, right, I mean with us. No, it's a sort of what is that dramatic irony, in the sense that we know what's going on. The person doesn't how we see. Uh, big tech actually like this is. This is big tech pushing for this future, because think about the bella ramsey ads from apple um, where the actor is but sort of starts to walk into a room, see somebody that they're supposed to know, backs up and then asks Siri, you know who's this person, what's where, do I know them from? That kind of thing. Siri tells them and then they're able to go on and have the conversation. It's being pushed as a cool future feature, a feature that you would want to have, an option that you'd want to have. But while that is that, that the way that that's done is through somebody's own. You know data that's available, where it's looking at your location, that you've shared, that you've done. You know you've opted to blah, blah, blah. Having it be the other way around is chilling always come back to when it comes to this idea.

There is a part of me that is very aware of the fact that we are all given, when we sign up for a service, the opportunity to read through terms and conditions. Yeah, and we are all given the choice of whether we sign up for a service or not. Right, and we do have a responsibility to understand or choose, not to what we're giving when we make those choices. And it's tough because the sort of soft, humanitarian side of me which is a much larger side of me is, you know, railing against the machine here and saying, you know people need to be, we need to do a better job of telling people what's going on, have them be more aware of it. But there is a small part of me that, I guess, tries to have empathy for literally everything, and so I have empathy for the company in that situation. And I say the company's doing their due diligence in having it be in the fine print that this is the situation and that your data is available.

And I've noticed, as many people do, that it does oftentimes take object lessons for humans to learn not to do something or to be better about something. And so maybe these are object lessons that have to happen, this thing where it's like let me show you what it means because you've decided not to read through the fine print and you've just said yes, yes, yes, yes to everything. You need to see that to understand the impact. I'm curious, abrar, understand the impact. I'm curious, abraar, anecdotally, in people that you've talked to that are outside of tech. How often do you find that they have a deep concern for their privacy? And if you do take the opportunity to kind of explain, hey, when you're watching TV on this TV that you bought, that was a budget TV it's tracking what you're watching. I want to hear what's been your experience in that kind of thing over time. And if people are concerned about this or they're just kind of like, well, you know, it is what it is.

1:00:58 - Abrar Al-Heeti
I think that's exactly the problem. Well, you know, it is what it is. I think that's exactly the problem. I think we get really excited about new technologies and we get really excited about oh, this is a free service If I can sign up for any of these social media platforms, this is a discounted product that I can take advantage of. And then you realize there's always a price, right, it's not. You know, there's no such thing as a free lunch, there's no such thing as free tech, and so you kind of learn a little bit too late what the cost is.

So we see people who do kind of have to wait for something like this to come out, where they're like oh, this is quite serious, and it's not like it just sprung up overnight. Right, we've been building up to this for so long, these platforms have been collecting our data for so long, and so then you get a piece of tech like this, and it's not just about the glasses right, people are concerned about what glasses can record and capture, but it's about everything that it's able to access and what that bigger picture looks like. So, yeah, I think we who are in this space, whether we're reading or writing these tech stories, we know what the privacy implications are, we understand the dangers, but we still engage in it, right, and so imagine if you don't necessarily know um everything that's at stake and um and yeah, I think a lot of it is.

1:02:23 - Mikah Sargent
Just, I celebrate this, this use of the technology as a means, because, very clearly, you know they're not making it available to people. They're literally showing you look at this things you can do with it. It's not great that this can, this can happen. And so, going back to the question that I asked you, answering it for myself, I will say that I have seen, which I've been happy about, an overall increase in awareness of one's sort of data value and one's privacy, and I like that. That has happened, yeah, but it only reaches to a certain extent. And that's been the fascinating thing for me is, while I have seen more people showing concern about an app, say, tracking them I specifically brought up the TV example because that is so far down the line for people that, again, anecdotally, I've explained that in the past and they just that's just one that they don't really care about. And there are a few out there where it's like, oh, that's doing that, that's fine, that you know, that makes sense and it's it's not to the level, uh, that we have and the awareness.

I think that we and by we I mean those who study this stuff every day and who do, in different ways, read through terms and conditions and understand the implications therein. It's not there yet and I'm curious if it will get there or if, because I almost saw there used to be a whole heck of a lot of apathy about privacy and especially the phrase you would always hear They'd always kind of joke I don't care if the CIA is looking at my stuff, I'm not doing anything wrong. If somebody wants to look at my stuff, I mean, I'm just boring, so who cares? That was the refrain I used to have, that, that used to be my refrain. And then I remember the day that it changed, because a good friend of mine convinced me that that apathy was only making things worse.

And then I, you know, so I became kind of hypersensitive to it, at least in awareness of it, and I saw a decrease in that apathy, an increase in that concern. But I'm wondering if it's tanking again a little bit, because it can feel a little overwhelming, right when you just were at a place where you kind of have to accept and this is the thing that I kind of struggle with saying that but do you, you almost do have to accept that it's going to be out there. We're all having not just our phones now but also our little lenses on our face and all this other stuff, and given that that technology is there, you either exist in society or you don't, but we don't really have the option to not exist in society, or you don't, but we don't really have the option to not exist in society. So, yeah, it's just kind of there.

1:05:28 - Abrar Al-Heeti
I do think we've kind of grown numb to it, because think about all the times that we hear about all of our most personal data being leaked right by whatever it is, whether it's a corporation or whether it's a product that has been violating our privacy.

I do think it's so hard to still feel shocked or angry by it when it happens all the time, and so that's just kind of a disappointing reality. And then, on the flip side of it, these companies clearly create a product that is valuable for people, and so they're not going to walk away from it if it's a tool that helps them connect with people or makes their lives easier, and it's just a really unfortunate cost that you have to pay. I mean, that's the price that you pay for something that you want to use and find it difficult to remove yourself from. And even if you don't, even if you're not on social media again, there are ways that your information is going to be compromised anyway. It is so hard. Privacy I don't think it exists anymore and I just think it's a tough pill to swallow, but that's kind of where we're at right now.

1:06:26 - Mikah Sargent
Thank you again, Abrar Al-Heeti, for all of your great work this year. A couple of interviews to round things out. The inimitable Leah Nyland of Bloomberg joined me twice this year to talk about the Google antitrust case. This was my first conversation with Leah Nyland and I learned so much and was honored and overjoyed to have her back again later to kind of give us an update. So I plan to see her again in 2025. Let's just kick things off with a big question. Can you help us? Can you explain the significance of the DOJ considering a breakup of Google? How does this compare to past antitrust actions against tech companies? And, you know, is this novel or have we seen something like this before?

1:07:16 - Leah Nylen
Yeah, so this is a pretty historic recommendation. The Justice Department hasn't recommended breaking up a tech company since Microsoft, which happened 25 years ago, and the US hasn't actually broken up a company since 1984, when they did it with AT&T. So this would be something that hasn't happened in 40 years, if it actually does go through.

1:07:38 - Mikah Sargent
Wow, okay, yeah, that is huge. That is huge, I'm sure, for you. Actually, I want to ask you, as an antitrust reporter, were you waiting with a bated breath, just wondering what was going to happen next? How was that experience for you in hearing about a breakup? Was it kind of like, you know sound the alarms, light the lantern, something wild is happening. What was that like?

1:08:06 - Leah Nylen
Yeah, it was pretty fun actually, you know, because we're the antitrust nerds, where there are very few of us but we're mighty. But this is like a historic event for us, as I said. You know, this hasn't happened in my lifetime, so it's sort of like a once in a-lifetime event, especially, you know, because they haven't done it in so long. As I said, they tried with Microsoft and they ended up and eventually accepting a settlement that didn't require a breakup. So, you know, we haven't seen this in a very long time.

1:08:36 - Mikah Sargent
Wow. So in the article you talk about how there are multiple potential remedies rather than going the way of a breakup like couples counseling. What are some of the other options that the DOJ is considering and how would those impact Google's business versus something like an actual breakup of the company?

1:08:59 - Leah Nylen
Yeah, so the Justice Department's filing. They described it as sort of like a menu of options that they're giving the judge, who will ultimately make the decision as to what to do. So the breakup is sort of like the on the one side, the most extreme. In the middle we have another one that they had recommended, and that would be to require Google to sort of share its index, which means sharing all of the data that it collects on the Internet to help make its search results and also help make some of its AI products. And the Justice Department says that they think that is a good option because it would help rival search engines, but it would also help small companies that are trying to get started in the AI business, because, you know, one of the big things that hampers folks is getting access to the data to build the models, and this would sort of help with that.

And then they had a couple more options that were more minor. Of course those are the ones that Google has said that they really like, and that's just to sort of unwind a couple of these contracts that they had with other companies, Like they were paying Apple and Samsung and some other companies billions of dollars a year to make Google search the default on web browsers and phones. And so Google says you know, we can just get rid of that requirement and sort of move from there. The Justice Department doesn't just want to do that because they say you know, that was the illegal conduct, we're going to get rid of that. But we also need to do something to sort of remedy what has allowed Google to maintain or to gain its dominance and then maintain it over the years.

1:10:32 - Mikah Sargent
Okay, yeah, this is interesting that there are several different options here, but that there's. It's cool, I guess, to see that it's not just about where we've seen kind of, I feel, rulings. In the past it's been about slaps on the wrist or bigger slaps on the wrist I don't know what you would call those, but breaks of the wrist I guess. But this is about not just sort of you know, making a company behave by by threatening it with something, but about actually getting at the root of something and pulling it apart, figuring out what needs to change so that it doesn't continue to happen or happen again. And you did write that Judge Mehta ruled Google broke antitrust laws in online search and search text ads markets. Can you kind of talk about the key issues in those findings in particular?

1:11:28 - Leah Nylen
Yeah. So search everyone knows what search is. You go to the thing you type in your thing. It gives you the results. The search text ads are the ones that sort of appear at the top of the page.

Advertisers pay billions of dollars a year for those advertisements, and Judge Mehta found that Google has sort of carte blanche over how much it charges those advertisers because they really feel that there's no other option. You know, when you go to a search engine you are oftentimes looking for information. You might be looking for information about a specific purchase, so advertisers really want to get in front of you right then because you are thinking about buying something. So advertisers really want to get in front of you right then because you are thinking about buying something and there's really, in their view, sort of no substitute for these advertisements, and that you know this is actually Google's core business. This is where it makes two-thirds of the money that fund the company.

So the Justice Department was pretty happy with the finding that not only had they monopolized search, they had monopolized these ads, because if you want to change a company's business, you really do need to get at where it's getting its money from. And so some of the suggestions that they made in this also do relate to the advertising market. They made some specific recommendations there, such as giving advertisers a lot more information about where it is that their ads are actually appearing. When advertisers go and buy stuff from Google, google doesn't actually give them like a detailed list of you appeared in this many searches and like this is what people were searching for. They give them like sort of an overview of where they were spending the money. But the Justice Department said you really should be like giving people a receipt, like we appeared in this many searches and things like that, and allow them more controls over which ads they appear in and which they don't. So there are some proposals in there that would get towards this side of the market as well.

1:13:19 - Mikah Sargent
Interesting. Now, this isn't Google's only set of ongoing issues when it comes to anti-trust actions. So tell us about kind of the other stuff that's going on With your experience. Do you think this is going to have an impact on those other cases, like the one it has with Epic Games and the display ads case? That's a completely separate thing.

1:13:43 - Leah Nylen
Yeah, so in the US there are the three sort of big cases. The one is this one that's focused on search. The second one is the one that you mentioned involving Epic Games and a bunch of state attorneys general who also sued over related things, and that one involves Google's Play Store. In that case, a jury found that Google was illegally monopolizing the market for Android apps because it had been paying manufacturers to ensure that Google Play was the only store that was pre-installed on devices. And then they were also paying a lot of major developers to ensure that they would only debut really important games in the Play Store as opposed to other places, and they were paying, again billions of dollars to various companies for this privilege. This is a smaller side of Google's business. It's only $14 billion a year. It's sort of funny when you say only and it's still that large amount of money. So in that case, the judge has already made his ruling. He has now forbidden Google from entering into these kinds of deals for the next three years, and he has also ruled that Google has to allow third-party app stores to be downloaded through Google Play. So sometime in the near future you may be able to download other app stores within Google Play onto your Android device to make it easier for you to get apps without going through Google Play's billing. So you might be able to start paying for things using things like Square or PayPal, so that the merchant then doesn't have to pay the 30% fee that Google charges them to use the Google Play billing services. So Google has already said it's going to appeal that one. That one is largely done.

The other one that we mentioned involves ad tech. So ad tech is this sort of complicated technology that you as a consumer never see, but it helps buy, sell and serve all of the display ads that you see across the web. So those are like the banner ads that you see when you're reading a news or a blog or a sports site. Google has also, according to the Justice Department, monopolized the market for that kind of technology through a series of acquisitions and conduct over the years. That case just had a three-week trial that wrapped up two weeks ago. It's going to have closing arguments in November and then the judge said that she will rule before the end of the year. That's another one where the Justice Department has actually explicitly said that they want to break up. So if they win. They have said that they think that Google should be required to sell off some of the tools that it has bought and owns that relate to online display ads.

1:16:25 - Mikah Sargent
Yeah, yeah, now in it you note that, as you would imagine, google does plan to appeal once everything's kind of figured out. It's just that Google's like look, we're already ready. So, based on your reporting, what do you think this is going to look like timeline-wise? Will it be reminiscent of what we saw with Microsoft? Do you think it's going to be a whole new ballgame? Is there a shift kind of in the overall way that the justice system looks at big tech companies? What's your crystal ball here?

1:17:04 - Leah Nylen
So the appeals process, as I said, it's starting now in Epic. It won't start in the Google process until the remedy is imposed which the judge says he's going to impose a remedy by next August. So that's sort of our timeline for when this search case will be done and then that appeal starts. When this search case will be done and then that appeal starts. Appeals generally take anywhere from one to two years Depends on how busy that court is at that point in time. And then you know there's always the possibility of a Supreme Court appeal.

But the Supreme Court really doesn't take antitrust cases very often. They didn't take the Microsoft case. You know they don't take these very often because they're boring from their point of view. So you know it could go to the Supreme Court or it could just go back. So you know Google has to wait to the end of the whole process, both the liability and the remedies phase, and then it gets to appeal. It is going to, of course, ask that it not be required to do anything while that appeal is pending. The Justice Department is probably going to oppose that and say well, you should at least have to change, make a couple changes while that is pending, in particular in the search case, because some of these changes have already started taking place in Europe. So they feel that you know, in their view you're already doing it elsewhere. You might as well start doing it here. So we'll sort of see. Google has been emphasizing that we're still years away from a resolution, but we're probably more like two to two and a half years out.

1:18:29 - Mikah Sargent
Understood. And then one more kind of crystal ball question here, and really every time I bring up a question like this there's a little bit of understandable hedging involved. But it's an honor to get to talk to people who are very versed in the work that they do specifically, and so more. I'm asking from your perspective as an antitrust reporter, as someone who focuses on this stuff regularly, given the complexity of Google's business, what challenges do you foresee in implementing a potential breakup or other structural changes to the company if one of these other remedies plays out and the appeals process doesn't work out in Google's favor?

1:19:15 - Leah Nylen
Yeah, I mean Google has already actually started talking about some of the things that it thinks would be sort of technical challenges if it would be required to be broken up. For example, you know Android and all of the App Store and stuff. These are like software but they run on Google's machines, so if it's required to divest it, you're going to need somebody else who has a lot of servers that can devote to this, and there actually aren't that many other companies in the world that have the sort of computing power that would be necessary to sort of host all of this stuff. I mean, you have Microsoft, you have Amazon, but do we really want to be giving them, you know, this other major business? They've also, you know, in some cases been accused of being monopolies as well.

So you, it's a little bit complicated about who might be able to buy this, or if it were spun off into a new company, like, how much of Google's existing stuff does it get to take with it? That is, you know, pretty hard to figure out. I mean, the other thing is, you know we were talking yesterday with analysts and you know they said Google's future is a lot in AI. The Justice Department is looking at putting some of the limitations in what they can develop with AI, because so much of the data underlying its AI models came from search and their argument is if you monopolize this market, you shouldn't really get to take the fruits of that illegal behavior and move it to a new place. So it could be, you know, pretty, I guess, catastrophic for Google, depending on what the judge rules, if he does place limits on their development of AI.

1:20:56 - Mikah Sargent
I was hoping that in talking about this credential portability, we can talk about breaking down the kind of key problem that these new specifications are trying to solve. Why has credential migration between providers historically been such a challenge?

1:21:13 - Nick Steele
Yeah, oh yeah. No, David, take it away.

1:21:16 - David Turner
Oh well, the problem is that everybody keeps it in their own ecosystem and while that works really great within the ecosystem, there are a lot of challenges when you try to shift ecosystems. If you want to go, say, between Apple and Android, or even if you have reason to move between different PASCII providers, there are barriers to doing that and the solutions that are out there tend to be very insecure. You basically save things to text files to your local machine, and that's a really bad thing to do with important credentials, and on mobile devices there's largely no way to do that at all. So it's kind of solving two parts of the problem making it more the capabilities more broadly available to the industry and, at the same time, making it a very secure way to exchange the data.

1:22:09 - Nick Steele
Yeah, right now the kind of default across the industry is to export a CSV. It's unencrypted, unformatted. It's generally a pain in the butt, not just for users but for companies in the credential management space, to marshal and format these files and sometimes that data gets lost. So there's no guarantees right now that your credentials, when they move from A to B, when you get a new phone and potentially a new credential manager say, you're moving from an iPhone to an Android device that those credentials are going to all be able to be imported or exported properly.

Additionally, credential exchange is not just about import-export migration. It's about shareability. If you have one password today and someone else has Dashlane and you want to share a credential, there's no great way of doing that. So if I want to share something with my mom, who's actually a Dashlane user what? Yeah, well, I use a few, so I'm not picking favorites. But the process is she has to go either download 1Password or the credential manager that I'm trying to share with, or she needs to copy-paste the credential, or I need to copy it to her in a separate chat, which undermines the whole reason for having a manager in the first place.

1:23:31 - Mikah Sargent
Now something that stood out to me. The specifications mentioned 12 billion with a B online accounts that can now be accessed with passkeys. You know, the narrative seems to be that there are still very few accounts out there that support passkeys, but that's quite a few. So how significant is this standardization effort for accelerating actual passkey adoption across the industry?

1:24:00 - David Turner
It's critical. I mean, when you look at Google's already announced 800 million users have access to pass keys, and Microsoft's rolled it out to all their MSN users and Amazon's announced like 175 million people. When you start adding up, that's three. And then Apple and their ecosystem is about 1 and a half billion users. That's you know already. You're at two and a half three billion, and that's just three accounts.

You know, everyone always talks about the number of accounts they've got there, and an effort like this requires standardization, because if you well you know, if you only have one, of course it's useless. But even if you just have two vendors involved, it still has little value. What you really need is it to be as available and as well integrated as the rest of the ecosystem, because that then removes the barriers to adoption. You know, anytime you're trying to use a new technology, there's the technology itself and then there's how do you use? Use the technology, and the ability to move the passwords around is an important part of this whole ecosystem, and so having a secure way to do that is a really, really important key to getting this more broadly deployed.

1:25:12 - Mikah Sargent
Something a little nerdy. The specifications address both secure protocol CXP, and standardized format CXF. I was hoping you could explain why both components are necessary for this credential exchange to work as it needs to.

1:25:28 - Nick Steele
Yeah, absolutely. The credential exchange protocol just defines how we can securely and privately move credentials from A to B. But what's really critical is that same thing I was talking about earlier on is we all need to be able to speak the same language once the credentials move, because we may get dropped data. This is really critical, and kind of new to the space of being able to move credentials is that we're going to start defining formats for how credentials should look in code or really just in JSON, for how they should be moved, stored and securely backed up. This doesn't just help with exchange itself and sharing, but for the long-term format and storage of credentials and data related to identity, this is necessary. There's other standards, like SCIM2, that exist for managing and formatting identity information about users. The same needs to exist for credentials.

1:26:30 - Mikah Sargent
Now you look at the list of participating companies and it's, you know, apple, google password managers, like one password, which is the password manager I use Bitwarden. How difficult was it to get these companies on board, I guess more. What I'm asking is did everybody kind of go? This is something that we really just need to make happen, and so it was just easy to kind of get everybody on board, or was it a little bit like pulling teeth? If you can answer that, yeah, I can.

1:27:08 - David Turner
I can talk about that one. So, um, pretty much everyone in the press release is already a member of the fight of alliance, where we define, you know, the infrastructure and such for passkeys, and so we already have all of those people having conversations about making pass keys work. They're already invested in the ecosystem and so it wasn't't a big leap and it was a pretty clear step to say, yeah, we need to be able to move this stuff around. So for the people that we were already engaged with, who were already members, it actually wasn't a big step because, as I said, they're already doing this work. They already know what their customers need and, quite frankly, are asking for. And so, you know, yeah, there's always battles about what's in the spec and you know, how does that fit within an ecosystem and so on. But at the conceptual level, we had pretty much everyone on board pretty quickly and it's been a very positive collaboration, which is always great to see.

1:28:01 - Nick Steele
Yeah, I think that's one of the things I love about working in the FIDO Alliance the most is, I've been involved in FIDO for eight years and I've been in standards for 10 as a whole across a lot of different bodies, but it's a really collaborative nature.

We don't come at it from a competitive angle and we're there to make credentials and passkeysASCIs, specifically in the Fido Alliance, work for users. One of the things that I feel like we managed to really get ahead of is I think early on this year you saw a lot of press come out saying that PASCIs are going to be a walled garden. They're a way to lock people into platforms and keep people from moving off of where they keep their credentials, and that's really not the case. We just really wanted there to be a secure way for credentials to be moved because you know, unlike pass keys, if you lose the private key to these credentials, then that's a higher security concern, because we're putting a lot more trust in these types of credentials than we are passwords in the past, putting a lot more trust in these types of credentials than we are passwords in the past, definitely Now.

1:29:08 - Mikah Sargent
the protocol does emphasize secure credential exchange in both online and offline scenarios. I think the question is a little obvious, but at the same time I'd love to hear these nerdy details of why it was important to support different network conditions. You know, because you've got some services that use the cloud for a lot of things, but then there may be times when you don't. So what led to the choice there?

1:29:30 - Nick Steele
I'm really glad you called this out. I was saying the day before the call, I was like this is a really nerdy question. I'm really glad to get like the good ones. Yeah, so we really wanted to make credential exchange work in a variety of scenarios. Right now, you know, as we said, right, the default is a CSV. That you know you have to kind of hope that users delete. You know, make sure that users are going to do this securely.

Well, not only do you need to support these, like consumer cases where you know I'm just trying to move to a new provider, but you need to support these business cases, these more high-assurance cases where you know being online may not be available, you know you may be in a secure environment or you may be trying to move credentials from you know, your laptop in your office to a rack in your data center and there may be a firewall in between or some air gap where you need to be able to move those credentials with different network conditions or in cases where they can't reach out to the same services and backend.

One of the things that we really explored in Credential Exchange and are going to continue to develop is these enterprise specific flows, these high assurance, high security flows, where the business actually has a lot more authority around the movement of credentials and could potentially apply things like policy or authorization certificates and signing that could say that, yeah, this user is allowed to move these credentials and only these types of credentials from this provider to this provider, and then, in the case of these certificates being there, being online or offline may be an optional thing, but the business still has the ability to authorize this movement, and that's something that's brand new and, I think, pretty innovative in the space that we worked on a lot in FIDO.

1:31:30 - Mikah Sargent
One of the key features, of course, is the ability to securely move pass keys between providers. And I think from the outside, looking in, you hear one company supports pass keys, another company supports pass keys, a password management platform supports passkeys. It seems like if we all have the same little key, then moving my key from this place to this place and that place to that place should be easy. But there were clearly and are clearly challenges. What were the specific challenges, or maybe the biggest one or the biggest few, that needed to be addressed for this credential exchange to be as simple as I imagine you all intend to make it?

1:32:21 - Nick Steele
Yeah, I think.

1:32:21 - David Turner
Oh, sorry, David, go for it, yeah, oh no, I'm just gonna say that, that it in my mind, it and it's not. This is not a geeky answer, but it boils down to being able to do it securely. Right that right off the top. If we're going to move stuff between two points you know important stuff it's got to be done in a very secure, trusted way.

The second is the data formats vary wildly. You try and move stuff between password managers today and you get nominal levels of success because the way they define all the elements of a password versus, say, credit card information, it's not the same. You get some of the data coming over, you miss pieces, and so the value of having agreed upon data structures is actually a huge value as well, so that we get a much better quality of information exchange and it's easier to say at a whole level I recognize this credential type or I don't, in which case you can be really clear that, all right, the ones I recognized, I brought them in, but here are the ones that I didn't, and so you can give the end user a clear indication of what worked and what didn't work, whereas today, yeah, some of the stuff might roll in and you really don't know what did and what didn't.

1:33:33 - Nick Steele
Yeah, behind the scenes, on a lot of these credential managers, the plumbing is very different, is very different. So being able to have this protocol in place that is going to standardize how we communicate these credentials was the biggest first step. I think there's still a lot of work being done behind the scenes and at 1Password, dashlane, bitwarden and a lot of these orgs that we're working directly with, and we'll have a beta rollout of CXP with shortly of just like making sure that we're able to, behind the scenes, get these credentials in the format they need to be exchanged and work like we say they're going to work.

1:34:13 - Mikah Sargent
Understood. Now let's talk about the future. You know these specifications at least. When I heard that this was coming through the pipeline, I thought, okay, this is it. Now I can get excited about past keys without seeming like I am, I don't know, thinking the earth is flat in the sense that people will actually believe me and be excited that this is the future. But I wanted to ask you both, you know, do you think this is the last big hurdle for making passkeys a more common form of login or other type of authorization authentication? Excuse me, other type of authorization authentication? Excuse me, or do you still have some boxes that you're looking to check in terms of getting you know, my cousins and my uncles and everybody else who's not super steeped in technology on board with wanting to use them?

1:35:19 - David Turner
From a technology standpoint it's a big one Because that lock in notion or fear of lock in was, as Nick said earlier in the year, being discussed quite a bit out in the wild. And but it's. What's interesting is awareness is a big part of it too. It's not just a technology. They're not just technology gaps, there's just awareness gaps. People see passkey. They don't know what that is. They don't know if they should trust it. The notion of doing things without a password on the surface sounds scary to people, even though it's more secure. So it's it. The deployment is dependent on a variety of factors, not just the technical ones, but absolutely is a big, big step forward on the technical side.

1:36:03 - Nick Steele
Yeah, I think the I mean with regards to CXP right, the first step is allowing for passkey portability and we're actually, you know, going to come out including Bitwarden, dashlane, google 1Password are going to work on supporting passkeys, passwords and some other information, I think, addresses credit cards pretty soon. We want to be able to support a broader range of credentials and we see credential exchange protocol really being the first step in a broader conversation about where credentials are headed, especially as we start seeing more digital credentials and MDLs mobile drivers licenses become more widely available. We want to be able to support the movement and management of these credentials and allow for that user to have sovereignty over their credentials and making them more portable. So I think this is a huge leap for passkeys, especially, as David said, we're kind of moving away from that fear of lock-in. We're moving away from users feeling like they're going to, if they're moving from new devices or moving to new apps, that passkeys are going to be a risk to accessing things that they need.

We're just going to make them more available. They're going to be available across more credential managers and I think that's sort of a future that we're also planning for, I think, especially as we start talking about mobile driver's licenses and having digital wallets become more widely used. We're going to be entering sort of a future where you're not just going to have one password or Dashlane or one credential manager. You're going to have multiple sources of credentials and we're all going to need to talk to each other, and we see CXP as a way to enable that, which is like the FIDO Alliance. We're starting with pass keys, but we're seeing we're building for the future.

1:38:03 - David Turner
Yeah. So I had two things to that. At the consumer level, this notion of well, they tell you you're not supposed to do it, but everyone does sharing your streaming service account password with a family member. You know, at this point it's like copy it down or send in a text message and you still got to. You know, enter it in this way. If you can give someone and I won't name a particular service, cause I'll get mad If you you share the the account information with them. It's in their password manager and it will simply work.

So the usability of that scenario is actually much simpler. Um, it's the same with, you know, shared bank accounts. Um, some of my bank accounts, some accounts with my wife. We can only have one, right, well, so I have to share, we have to share. So, again, this makes it simpler and some of the Passkey password manager tools actually give you some control over that as well, so that that can be managed. In addition, you actually have regulatory requirements In the European Union. The EU, they're developing digital wallet infrastructure, if you will, and one of the key requirements is portability of data. That's a really big requirement in a lot of the EU. Cybersecurity and data areas is to make sure that end users aren't locked in to any one platform or any one product, and so this is the kind of tool that enables that capability again in a way that regulators will like it, as opposed to again doing it in an insecure and unreliable way.

1:39:37 - Nick Steele
I don't think this is like the end of the line, too for what needs to be done on pass keys by any means, and a lot of that work is not just being done in FIDO, it's being done in the W3C and the WebAuthn working group, which I'm also a part of and David's involved with. We've been working on this for seven or eight years and the work continues. We're due to publish the next version of WebAuthn and the WebAuthn API in the next month or so, and that's going to include a lot of improvements around usability of passkeys and the underlying APIs which is defined by WebAuthn in there. So we're really interested in handling scenarios for, I'd say, the more high security holdouts in the space, being able to make these more usable. Well, you know, usability for people is great. Usability for, you know, your bank is also important.

Banks and other high assurance you know payment service providers that experience a lot of fraud really want to make these credentials a strong, you know, a stronger replacement for passwords, not just for them but for their regulatory needs, as David mentioned.

So we're adding a lot of things to support better, I'd say, usability, better signaling between applications and your credential managers, being able to handle things like passive enrollment and registration of passkeys is something that's coming down the line.

So being able to go to a site, log in with your password and then having one password pop up and say hey, we saw you just logged in, do you want to enroll a passkey? And we can just seamlessly make a passkey right after login with this site. So we think that's going to be a huge step forward in terms of usability and adoption If we just make, if we make that process easier and easier to just roll over to something more secure. Because you know, right now I, a couple sites are have have a pretty great flow, but if you want to roll a passkey today, okay, well, you log in, you go to your security settings, you go to write add a passkey, you add the path, you name the passkey and it just becomes this certain amount of friction. And we really want to bring that friction down and make it just as seamless as registering an account with a password.

1:41:50 - Mikah Sargent
Absolutely Well. David Turner, nick Steele, I want to thank you both so much for taking the time to join me today to talk through this update. I'm very much looking forward to seeing how this progresses. Is there a place where folks should go to kind of stay up to date with what's making its way down the FIDO pipeline, as it were?

1:42:14 - David Turner
fidoalliance.org is our website and that's probably the best starting point. You know, from the top level. There'll be links down to both the PASC-E related documents themselves, but also credential exchange information too.

1:42:28 - Nick Steele
You can follow me at Kaiju on Blue Sky. I post there and have a couple of lists of other folks like Matthew Miller and Tim Capali who post a lot in the space, and we'll keep you updated on where the specs at.

1:42:42 - David Turner
Actually, one other quick thing to add is that there are a lot of open source initiatives in the same space trying to do this kind of transfer, and so we've taken a unique step with FIDO and set up a repo in GitHub specifically to get feedback on these specifications. So there is a place, a direct place, to provide comments. It's a publicly available repo and again, it's on our credential exchange download page. I don't remember off the top of my head the name of the repo, but if anyone has been looking at the specs and they have comments or feedback, we do have a GitHub repo for that.

1:43:21 - Mikah Sargent
And then, lastly, I was pleased to be joined by both Nick Steele and David Turner from the Fido Alliance, who came on to talk about how passkeys are finally getting portable. Will 2025 be the year of passkeys? You'll have to tune into Tech News Weekly all this year to find out. Well, next year. Thank you so much for this year of Tech News Weekly and I will catch you again on the other side of the calendar. Bye, everyone.