Skip to content
PodSubs
ShowsPeopleGuestsJobsNewsEvents
More
BlogPodCred
Menu
ShowsPeopleGuestsJobsNewsEventsBlogPodCredNetworksSign in
Join nowSign in
  • Shows
  • People
  • Jobs
  • News
  • More
    • Guests
    • Events
    • Blog
    • PodCred
    • Networks
    Join nowSign in

Future Media Podcast

Future Media Podcast artwork

Tech's legal nemesis opens up on threat to Google's shield

Jun 30, 202651 minAudio
ShareXLinkedInFacebookWhatsAppEmail

Thirty years ago, a now dead US judge with a pretty patchy past created a legal precedent that stashed a timebomb under today’s internet. It lit a 30-year fuse that’s blowing up today. Seriously, this story is so utterly nuts you couldn’t make it up, so I’m being joined by one of Europe’s top anti-Big Tech lawyers Thomas Höppner to explain it. But before we get there, please join me in welcoming an influx of new subs including the architect of Europe's GDPR, an exec editor at Bloomberg, the Washington Post’s platform team, a product guru from India’s NDTV, the marketing crew at Zendesk, Apple in LA, a Pulitzer Prize-winning photographer and Internews Ukraine in Kyiv among many others… Привіт and please share with your friends… 👋 Now back to the main event, and it begins with The Wolf of Wall Street. You know the guy. The drug-addled chancer played by Leonardo DiCaprio in the movie. Jordan Belfort was his real name, and by a truly bizarre twist of fate, he’s now the epicentre of an emerging effort that threatens to topple Google. Way back in 1994 - when Boyz II Men topped the Billboard 100 - Belfort was deep in his drug addiction, consuming Quaaludes by the handful. His pump-and-dump scam Stratton Oakmont had been in regulators’ sights for years and beginning to fail when two posts appeared on an online bulletin board. They alleged it was “major criminal fraud”, “100% criminal fraud” and a “cult of brokers who either lie for a living or get fired”. This was the era of dial-up and the comments were anonymously posted on the popular Prodigy Money Talk board. Stratton sued for defamation. Prodigy responded that it would fight the case and prove the posts were true. Facing a trial he didn’t want, in the midst of his addiction and the chaos swirling around Stratton, Belfort agreed to settle. That meant cutting a deal. Prodigy agreed to an apology and a retraction. Belfort accepted, and both sides asked the court to wipe the ruling from the books. Only the judge refused, and that set in stone the precedent that because Prodigy moderated its message boards, and this comment got through, it was liable. And because Prodigy never got to prove the truth, and it was never tested on appeal - it has remained the precedent through to today. That legal mess it spawned is the foundation stone of Section 230 and its national equivalents around the globe. And it’s protected Meta, Google and others from monetising terrorism and other abhorrent stuff without any legal risk ever since. Yay, what a win for the world 🥳 And it’s down to this guy (not Leo, the other one…) Big Tech has driven a coach and horses through Section 230 to fast track their path to publish hate, suicide, addiction, murder and child abuse to drive their vast profits. It has led to the near collapse of journalism, immunity to harm kids, the devaluation of trademarks, a corrupt market based on surveillance and the disinformation dumpsterfire that’s today’s web. A poor judge and a crook busted by the FBI created the law that has made the bad guys rich and the good guys go bust. But… Something’s happened. There’s been a disturbance in the force. A German court has done something quite extraordinary which could reset the balance. And after months of planning, I am delighted to finally welcome one of my heroes - high-flying European legal eagle Thomas Höppner - to explain what happens next. He’s not like most lawyers. He’s a scholar-litigator, who is not just dangerous in court but he’s also a professor of business and IP law. 🍿 Sit down, because I’m not even half done. 📚 He has a PhD in competition law and has published six books and 100 articles. 🧠 He’s represented publishers, been an expert witness at the European Parliament, the Bundestag and Germany’s Monopolies Commission. 🎓 He has spent 15 years constructing the legal theories to undermine Big Tech’s legal protections and is now helping courts adopt them. 💰💰💰 He’s acted for victims and helped extract roughly €9 billion in fines from Google, and he’s still chasing €6 billion in damages on top. If you’re a publisher, you probably want to have him on speed dial. 🤙 Let me know if you want an intro. Thomas was a partner at Hausfeld for a decade, before jumping to Geradin Partners last year - the go-to law firm for complex antitrust and tech regulation. His boss Damien Geradin was on my pod last year predicting Big Tech was facing “a domino effect” and “the idea they can escape forever is false”. So buckle in, because this newsletter might convince you that’s now happening… 🎁 Before we dive in, I have a special perk for paying subs. This is the first report from my new spin-off Future Media Intelligence. It includes data and exclusive interviews from the Washington Post, Condé Nast, Bauer and News Corp Australia on navigating the changing landscape. It’s published in partnership with FIPP WAN-IFRA and is free to their members and paying subs of Future Media. Get your copy by becoming a paid sub. The Big Story: Sex, lies and a poisonous precedent Section 230 and its national equivalents have shielded the bad guys for years. It’s been a golden ticket that turned the level playing field almost vertically in their favour. Then AI arrived, and Google suddenly found itself under unprecedented pressure. Shareholders began demanding ever larger profits just as search’s cashcow was coming under pressure and regulators were plotting to break its monopoly. You have $4 trillion at stake. Imagine you’re CEO Sundar Pichai. What would you do? For most, the answer would not be to deliberately break the search cashcow and replace it with AI Overviews, but that’s what he did. He did it for profit and knowing it would come at a cost. One of those was bypassing publishers and the open web - which is what my report above reveals. But now the greatest cost of all might be on Google itself. Because it’s overstepped. That’s what this chat is about. Google has enjoyed legal protection from Section 230 because it’s been a platform publishing blue links. Only now it publishes AI Overviews, that transforms it into a publisher - and oh, how the rules have changed. Thomas, welcome to the podcast. You’ll object, but it’s fair to say you’ve been more successful than almost anyone in Europe taking on Google over the past 20 years. You’ve led the legal fight on behalf of publishers and consumers against Big Tech platforms and I’ve wanted to have you on for ages. Thanks for joining us. Thomas: “I’m very glad to be on.” So, a court has just ruled that Google is a publisher. That doesn’t sound like a big deal on the face of it, but it removes a crucial legal shield. Section 230 is an American law that says Google is not responsible for what other people publish on its platform. It’s a dumb pipe, at least legally. Europe also has its own version of this law, and Google has relied on it since it was founded. Only, it’s coming to an end. Thomas: “Yes, I think this is very fair to say, and to be honest, it comes with a bit of a history. “While it’s true that for a very long time, there has been this presumption that if you run a platform, you shouldn’t be liable for the content you share. “There was this presumption that if you merely store third party content and make it available for use on your platform, you shouldn’t be required to verify the legality of it. “You are serving a purpose. You have an important role, and that shouldn’t be made more difficult than necessary. “The liability should remain with the content creator, unless it meets some of the exceptions, such as having to take down illegal content once you’ve been notified about its existence. “Under European law, this was called the notice and take down mechanism, and that has worked for a while. “But over the past few years, judges increasingly learned that we need more exceptions to it and we need to be stricter with the platforms.” But now we have this German judgment. Thomas: “Yes, a court said if you are no longer a neutral and passive platform, and you interfere in the diss… This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit rickysutton.substack.com/subscribe

Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.

Transcript

From the show. Tap a time to play from there.

SPEAKER_01Hello and welcome to the Future Media Podcast and today it's a cracker. 30 years ago, actually about 30 years and about two weeks ago and now long dead US judge with a very patchy reputation created a legal precedent that put a time bomb under the early web and it had a 30 year fuse because it's happening right now. And this story is so amazing you frankly couldn't make it up. So give me a couple of minutes just to set the scene.

SPEAKER_01So you've probably seen the movie The Wolf of Wall Street. It's the guy who was running a dodgy banking scheme, if you remember. And this whole story starts there. He sued a very early online bulletin board. It was called Prodigy. And Prodigy was a joint venture between IBM and the department store Sears, and its actual intent was to clean up the open web. Its promise was that it would moderate everything that people posted and bring

SPEAKER_01a sheriff to what was effectively a wild west of the web back then. But what happened was it missed a post about the guy, the main guy in the movie, Belfort the Wolf. And it alleged, this post alleged that he was a fraudster. Belfort's firm Stratton Oakmont dragged Prodigy into court and in front of a judge and the judge ruled that because Prodigy did the right thing by moderating content but missed this important post it was going to be liable for it.

SPEAKER_01Now, Belfort, who was soon afterwards, if you've watched the movie, you'll know, was revealed to be a shocking fraudster, knew he didn't have a leg to stand on in this case. And faced with prodigy threatening to prove the claim in the post was true, the Wolf of Wall Street settled it to bury it. And that meant that the ruling was never challenged on appeal. It just became law and in so doing it injected a poisonous precedent into the web because that was the beginnings, the basis of Section 230.

SPEAKER_01Now, most of you will know this, but I'll just remind you, Section 230 is an American law that says platforms are not responsible for what they publish because they're just dumb pipes. unless they moderate. If they moderate, they become a publisher and then they can become liable. Now Google and other tech firms have driven an absolute coach and horses through that law ever since, pushing it further and further to the point that it's effectively become a get out of jail free card.

SPEAKER_01Tech platforms have evaded liability using Section 230 in cases involving sex trafficking, aiding terrorists and harming kids. And all this because a dodgy judge passed a case on a complaint that turned out to be entirely true and made a bad decision 30 years ago. And that's had the repercussions of leading to a near collapse in journalism, the harms that crush kids on social media, and a marketplace where the bad guys get rich and the good guys go bust. And you can read a little bit about this in my new FIP report,

SPEAKER_01Discovery and Distribution for the AI Age, which I've got out at the moment. And that's what this is about today. We're just getting started because there's a legal shift underway. And today, for the first time on the Future Media podcast, I have the most senior and successful lawyer in Europe who's been taking on Google for nearly 20 years, leading the fight on behalf of publishers and consumers against

SPEAKER_01the big tech platform. So Thomas Hopner, thank you so, so much for joining us.

SPEAKER_00Thank you very much for the kind invitation and introduction.

SPEAKER_01Very glad to be on your podcast. I'm so thrilled to have you on. All right, so let me just set the scene before I'm going to let you loose because you've got so much to share. Google right now is under unprecedented pressure from dozens of angles. First of all, shareholders want it to boost profits with AI. The search tool it's relied on for years and years and years is under pressure, and it's been ruled a multiple monopolist in US court. So what do you do if you've got a $4 trillion monopoly that you want to protect?

SPEAKER_01Well, Google's answer has been a surprising one. It's to deliberately break search and replace it with, guess what, AI overviews. And it did it for profit. But that profit comes as a price. It's killing publishers who are seeing traffic fall by as much as 80%. But the greatest cost of all in the court of karma, as I've taken to calling it, might be on Google itself because it's overstepped. For all of these years, it's been protected because it just published Blue Links.

SPEAKER_01It was a platform. But now it publishes AI overviews and that makes it a publisher. And that changes the rules completely. In Munich, which is the one we're going to talk about most today, two publishers were defamed by an AI overviews answer. And the court ruled that Google was the publisher. Section 230 is an American law, but Europe has its own. and it's finally coming down against Google. Thomas, over to you. This is your world.

SPEAKER_01Do we have an opportunity here to feel optimistic that things might be about to start changing?

SPEAKER_00Yes, I think this is very fair to say. And to be honest, this comes with a bit of a history. While it is true that for a very long time, there was this presumption that if you run a platform, you shouldn't be liable for the content that you share. There was this presumption that if you merely store third party content and make it available for use on your platform, you shouldn't yourself be required to verify the legality of that content

SPEAKER_00that you share. You are serving a purpose. You are having an important role and that shouldn't be made more difficult than necessary. The liability should remain with the content providers unless some exceptions such as you know that there is illegal content on your platform, then you have to take it down. So under European law, this was called the notice and take down mechanism that worked for a while. But even over the last years,

SPEAKER_00judges increasingly learned that we need more exceptions to it and we need to be stricter with the platforms. And already before these German judgments, we've seen Decisions that say, well, if you are no longer a neutral passive platform, but you actively interfere in the process of the dissemination of information, then you lose your host provider privilege and an active role is deemed to exist,

SPEAKER_00for instance, where you as a platform actively promote certain type of content if you market specific offers if you contribute to the design of specific content and you process and enrich this content with your own information and structure and represented independently in these scenarios where you become an active part of it you lose your host provider privilege and in this setting the latest decisions

SPEAKER_00in Germany can be seen and they are in line with that in this particular case applicable obviously to AI overviews and I'm happy

SPEAKER_01to take you through that it's funny what you've just described when you were talking about building algorithms, promoting new information, adding new data to it, pretty much describes the patent that is AI overviews. That is almost the entire definition. It's the reason it exists is to do that. to answer the question there in the search query. And the only reason Google wants to do that while it will be telling you it's doing it for consumers and it is a good consumer product. But the reason they're really doing it is to hoard more advertising and keep traffic

SPEAKER_01on their own sites. And so how much of that? Why do you think Europe's waking up to this now? What do you think the courts are picking it up? I mean, you've been prosecuting on this kind of stuff since 2009. But why do you think the courts are waking up to it today?

SPEAKER_00Well, firstly, of course, AI overviews haven't been around for too long, but I think we will see more and more of those judgments. The better judges and also authorities understand how AI overviews actually work and how they are generated and what you see, the easier they will understand, look, This is something entirely else. This is not general search.

SPEAKER_00This has nothing to do with a neutral platform, disseminating fair party content. This is a publisher generating its own content, being in full control of what it publishes, how it publishes it, how it structures it. So, you know, there is no difference between that operator and publisher of a magazine, for instance, and accordingly, all those privileges that search engines enjoyed in the past need to fall away

SPEAKER_00because here the search engine is not really pursuing that objective of navigating users in an objective manner to relevant sources. Instead, it substitutes those sources and obviously becomes a publisher in itself as you rightly summarized. Coming back to your question, I think the awakening is basically the result of a better understanding what's going on. And I can't see that this will change because it's so crystal clear

SPEAKER_00that this system is not compatible with what we understood as a platform in the past.

SPEAKER_01so ground zero for this is Munich right so from what I've been reading this particular case with these two publishers maybe you can just explain what happened there please yes

SPEAKER_00so in simple terms someone asked the search engine Google is it correct that publisher XYZ has a dubious reputation and the AI overview answered yes this is correct publisher xyz has a dubious reputation they have several have been several reports that leech that they've done this and that bad things so basically an affirmative answer to

SPEAKER_00the question whether someone has a bad reputation that kind of answered it in a way yes you're absolutely right they are a bad company and it turns out that nothing of this was correct because even the references that the AI overview mentioned were either non-existent or they said something entirely different so nothing backed this statement yes they are dubious and pursuing dubious business practices entirely false and for that reason

SPEAKER_00the defamed publisher sued Google and said look this is defamation please stop it and I don't see how you get out of it and the court agreed obviously Google said as in the past look these are just kind of search results we are not liable for that go after those sources that mentioned that they are dubious practices but not after us the court was not having it at all

SPEAKER_00and I would like to explain a bit the reason so in basic terms the court said since Google introduced the artificial intelligence answers a overviews itself and offers those a overviews itself to end users it must also be held accountable for what these answers say as only Google itself has influence over

SPEAKER_00the content of its own answers and is in full control of all the algorithms that shape frame and disseminate these answers so in basic terms AI overviews are product of Google they are normal statements by Google itself that go beyond search results that merrily Link you to somewhere else that merely reflect what someone else has said.

SPEAKER_00This is own genuine content for which of course Google needs to be accountable, including for any hallucinations within. And that is really, in my view, a game changer because it brings to the legal sphere what technically is rather clear. If you only take third party sources as an input to generate your own output that is then independent of the sources and may come up

SPEAKER_00with something entirely else structured in your own way, adding information as it pleases, then this really has nothing to do with is a search result anymore. And accordingly, you as a provider of these answers need to be liable for anything it says. And that mechanism has, of course, far reaching consequences beyond the law of defamation. It also means that if we've got services that Impact now how ads are served how

SPEAKER_00the creators are designed how they are disseminated how the ads look and all these things wherever the platform actively gets into the process of generating content whether paid or unpaid they become liable for what they then present to the audience because they are in full control of that content And that is, of course, for many reasons, very important that the platforms can no longer hide behind the liability exceptions

SPEAKER_00and say, look, has nothing to do with us. If something on our platform is wrong, then, you know, go elsewhere, but leave us alone. And I think this is exactly the right response to AI generated answers, because it you know increases the responsibility the accountability and creates a strong incentive for AI providers to make sure that the output is correct

SPEAKER_00and they can only ensure that by making sure that the input is reliable and that in term of course increases the incentive to really team up and cooperate with reliable sources through proper licensing agreements

SPEAKER_01Yes, like publishers. There are so many angles in this. So let's let's break those down one by one. So the first thing that I think is really interesting is that I have started to see court cases popping up around the world. There's a big one here in Australia arguing that advertising shouldn't be hidden behind Section 230. So, you know, if somebody creates a fake ad that defames someone that they then publish on that platform, maybe even the ad has been created by the platform. Okay. And then that's then optimized and shared.

SPEAKER_01monetized distributed then that's still libelous even if it's an ad in the same way as if a publisher published a libelous ad and so you know that's one implication there's obviously the implication as well of holding effectively a company liable for an algorithm there was always this argument you can't say an algorithm is guilty of libel no But the company that generated it and created it for a specific reason can be, okay? It didn't spark into life. It was created by someone with a commercial reason to do it.

SPEAKER_01That's the second thing. I think the third thing for me as well, though, and I wrote about this a couple of days ago, I know you and I were chatting about this, is that it's not just civil liability either because if you're a publisher and you do something to derail a court case or you are Hallucinate something that has some sort of criminal impact, then there is also the possibility that editors, I've been one, we're always working under the risk of being held criminally liable for something

SPEAKER_01that we do because we're Reckless or because we've done it with a form of malice or because we've been reckless to ignore a court order or similar. You know, if you've now ruled that Google's a publisher, Google's responsible for every word it publishes civil or criminal, correct?

SPEAKER_00Yes, I agree with this analysis with all the three points that you said. Firstly, with ads that for which you need to have responsibility and should be liable. It was really annoying to see that, for instance, Google made money by showing ads for scams. And then they said, Well, you know, notice and take down, you can always inform us that there are scammy ads on our platforms.

SPEAKER_00And if we believe so we may take them down and practice they've been informed all the time but never took them down so that system never really worked so a direct liability as soon as the platform gets actively involved in either the creation or the dissemination or selection of those ads really would prevent some very bad ads and increase the quality of the internet and search in particular on your last point regarding liability

SPEAKER_00of individuals I think like criminal liability Indeed, I think we need to think about these consequences more. So far, businesses, large businesses seem to have thought, well, but if we earn more through that practice than any fines, we may ultimately have to pay or even damages by those that dare to come after us. As long as there's no genuine deterrent effect, they will simply continue doing so.

SPEAKER_00but the moment that individual criminal liability steps in the incentives shift and then on the second point what you said regarding the rule of algorithms they are this latest judgment of the European Court of Justice on the 16th of June so very new in the case Coyote system is particularly helpful they they said literally that an information service provider can no longer hide behind

SPEAKER_00a liability exception if they by means of an algorithm under their own control condition how content is distributed or rebroadcast so if you control the algorithm that determines which conduct content lands where then you are already having an active role in that and therefore

SPEAKER_00as an operator of those algorithms, you should be liable if false information is spread. So the tendency is clear. You can no longer hide behind algorithms that you control yourself and which you design in a manner that benefits your own commercial interests. So it's only consequential that if you do so, you bear responsibility for the outcome.

SPEAKER_01So I know you're very focused on Google as I have been, but for the past year and a bit, maybe less, eight, nine months, my focus has really shifted a little bit to Meta because Meta really is captured by what you're describing there with the algorithmic bit. I mean, that's been Meta's genius is Steve Smith Smith

SPEAKER_01it's not us we just built it and it's off doing its own thing that's fine but interestingly when you add AI to an algorithm then basically that says I'm not responsible for running you over in the road after I built the car caught the car and then gave the car an AI that just drove down the road willy-nilly you know we're getting to the point where what I think would be obvious to most people okay is that if you do something wrong you should be held liable for it but I want to ask this is the question We've kind of taught Silicon Valley over the last 25 years to move fast

SPEAKER_01and break things and we've told them that's okay and when it breaks just fix it and then we end up with a better product and people have genuinely accepted that. What we're now seeing is people saying getting to a better product isn't okay because of the actual damage that it does to get there. What do you think has changed? What's the trigger that's changed that do you think?

SPEAKER_00I think one of the reasons is this approach move fast and break things and then later on fix it has its conceptual flaws. The most obvious one being, sometimes if you break things, it's unfixable, it's broken forever, and you cannot undo the harm it has caused. And the risk with AI is particularly obvious that we move into a direction of where at some point it's too late to repair the point of no return and

SPEAKER_00if no security measures are implemented now they may be implemented at a point in time where we have already such powerful systems that maybe possibly beyond control and I think this is a threat that many have pointed out for quite some time and it is correct and I think what changed here is a better understanding for the devastating impact that AI can have and we've already witnessed companies

SPEAKER_00that say we don't even know ourselves how this outcome came about somehow our AI made it up But what exactly the reason for that is, we don't know. I think this is very varying if the company in control of its system argues that it can no longer control it and therefore it shouldn't be liable. In my view, this only confirms that we need stricter safeguards. and obviously there cannot be an exception such as our AI went rogue

SPEAKER_00and we have nothing to do with it and then basically stop the system the AI would be the natural fix but yes I think the awakening comes from the realization that simply telling companies now prevent this and that may be insufficient because they may not take it seriously enough and at some point they lose control.

SPEAKER_01I have a sheaf of papers here, it's 86 pages thick now, that I call the Big Tech Playbook. And every time I hear another one of those lines that they use, either in a court case or when they're attending a government briefing or when they're attending, they pop into Brussels every now and again and they do a glitzy dinner. and I've written them all down and I quite enjoy kind of ticking them off when I go to events. It's like a kind of bingo card. There you go. There's that one. But one of the things that we've been hearing a lot and increasingly so

SPEAKER_01and certainly from Sam Altman and some of the other AI CEOs is that what we're really looking for here is regulation, please. You need to regulate us. Come on. Now, I've heard that. I don't know how many years you've been in this. 2009. You know, I've heard that we want to be regulated narrative from pretty much every company at some point. where do you think do you think there is a need to continue to regulate or do you think it's the law and regulation or do you think it's just

SPEAKER_01the law you know what which one is going to work because one of them is kind of coercive I'm going to come up with some policies and say that you should do this but I'm probably going to build them with you and we're going to try and figure out how we can find some middle ground let's discuss it The law is very different. I've got 150 years of history that tells me that this isn't right and I'm going to deal with it. So I'm interested in where you think, you know, kind of the fix is and whether you think that this is just another page in the playbook.

SPEAKER_00See, when I read these statements, we need regulation. I sometimes even believe that because what we see is where there is no Regulation and I understand that a broader sense, including new laws that then govern how I can be rolled out. I think it's also a bit of a cry out to say, look, we are currently forced into a race to the bottom because some

SPEAKER_00of our competitors are not hesitating at all to make it worse and worse and worse. And we need to kind of adjust and apply the same devastating concepts they are using because we are afraid otherwise these really really bad guys disrupt us and then our business is under threat we need to defend ourselves so we need to copy paste these really really bad acts even though we know it's probably neither good for our users nor for our society

SPEAKER_00so what I read into this is actually please stop these really really bad acts from these even worse players than us so that we don't need to copy that to defend our commercial interest. So there is some kind of legitimacy into listening to that because, you know, they are responsible to their shareholders and if the shareholders expect copy these bad acts, otherwise you lose your business, they will have to do it even though they don't like it.

SPEAKER_00But apart from that, obviously we need both. We need a very strict application of existing laws where really affected companies go to court, say, look, I don't agree with this. It's up to you to decide that this is not acceptable, but at the same time, legislator need to really think intensively about how to get into, you know, the legal framework that sufficiently protects society from the harm

SPEAKER_00that we already see. and also to take down those really bad actors so that other ones that would be willing to follow the line of law don't feel the need to copy paste these really detrimental measures.

SPEAKER_01I was on stage at a legal conference recently and somebody you know at the end of the conference somebody said What do you think we need to do? You know, what are the priorities? And I said, Well, I think there are three things we need to focus on. We need to focus on the harms of the past. And there has to be compensation for people harmed in the past. And that may be core action, civil action, class action, whatever it might be. But we need to not forget that happened. I mean, we've probably got 10 or $15 billion worth of liability of losses sitting in there

SPEAKER_01that need to be recompensed. Okay. Then we've got the present and we need to think about what we're doing in the present. And probably for me, that's regulation. So obviously we've seen the antitrust efforts in the US, we've seen large fines and we've seen law changes and the new laws in Europe managing that. So we're starting to build the guardrails today. And then we have to look at the future, which is going to be about enforcing those guardrails and ensuring that everybody's kept on track when new businesses are growing,

SPEAKER_01governments are changing, perceptions are changing, you know, and so it's kind of past, present and future. Do you think that that's about right?

SPEAKER_00Yes, I totally agree with that. All three elements are essential. If you don't fix the past, the incentives to continue breaking the law will continue so companies need to realize that breaking the law comes with a price beyond pure fines where we know Europe for instance is very hesitant in imposing further fines in fear of trade retaliations so there needs

SPEAKER_00to be other deterrent effects and that can only be having to pay compensation for the harm that you've caused through your illegal conduct so it's essential that companies stand out and say look you broke the law be accountable for it this is the bill you need to to pay if they don't do that they don't just forsake their damages from for the past but they also We'd use the incentives to do it again and to take the mickey of them again in

SPEAKER_00the future because it had no consequences previously. So it's really important to do that. And then obviously there are also learnings from from those past infringements regarding how you need to set the framework for the future, because there is a repetitive rollbook. And if you see, okay, that was the legal strategy in the Past, it's likely they will try to adopt that because it worked so well in the future.

SPEAKER_00And therefore, there's also this benefit of, you know, learning what has worked and what caused significant harm in the past so that you can set up stronger legal guardrails for the future.

SPEAKER_01So I think most people are pretty across what's happened in the U.S. The antitrust trials in America brought by the Department of Justice and the U.S. government against Google showed Google was a monopoly in both search and search ads and also in ad tech. Yeah, I think when that started, I remember talking to publishers and saying to them, hey, this is a really big deal. And they were just like, that will just never happen.

SPEAKER_01No way that's going to happen. There was defeat before the first day of evidence. And then when the evidence did start, the message changed to it's a show trial. They've got to do it. Google's going to pay a small fine. It's just about making everybody feel a little bit better. And then when Google lost, they were dumbstruck. They couldn't believe it. And then the DOJ didn't hang back, did it? I spoke to Jonathan Cantor and I spoke to a whole load of other people at the DOJ and they were like, oh no, you should see our remedies package. We're talking about breaking them up.

SPEAKER_01We're talking about extracting Chrome. We're talking about breaking them into smaller entities so they're easier to regulate and manage in the future. And the publishers were like, that's never going to happen. And this time they turned out to be right. So on the search trial, it was a very soft remedy. It didn't really do anything. I know there's an appeal underway. I know that it's now being challenged, but we're still waiting on the ad tech. decision or the actual ad tech. Google's been found to be a monopoly, but what are they going to do about it? What's Judge Brinkham are going to do about it?

SPEAKER_01But my very convoluted question here is, is that how much is European policy led by American policy and how much is American policy led by European policy? How does it fit together? Because it's it's going these are global companies. And so you have to have a global response.

SPEAKER_00Yes, I totally agree with your analysis. It was really surprising how little trust American firms had in their own antitrust system a few years back. When I spoke to U.S. colleagues, they were like, you know, U.S. law, antitrust law is kind of dead. No way we can adopt similar cases as you've brought in Europe. We are simply not there. The Trinco judgments and so on and so forth are all in our path.

SPEAKER_00but we've learned that if you really dig into the system and you really understand what's happening and Jonathan Kanter fortunately was there at the right time in history He did a great job in bringing that case and winning that case on the liability front. Obviously, the timing was unfortunate for the remedy decision where Judge Meta got a bit nervous, wasn't certain what AI now actually means.

SPEAKER_00Is it a disruptor of search or will Google just continue on being a monopolist in this area? I believe if the judgment was made today he would have learned that actually Gemini is doing exactly the same as search has done in the past with the same tricks of tying and bundling and is now taking over ChatGPT in terms of users and maybe hopefully his judgment would have been differently

SPEAKER_00so yes this was disappointing.

SPEAKER_01Would you mind if I just jump in there and ask you a question? When that judgment came in, sorry, I've got to get the terminology right. When the trial happened, a lot of evidence was presented. He then wrote a very, very, very thoughtful ruling that Google was a monopolist. And it was very thoughtful. And when the trial was brought by the DOJ, AI wasn't really on the radar. There wasn't a single mention of AI in the original complaint from the DOJ.

SPEAKER_01And in his ruling, there were, I think, four or five mentions of AI, but it wasn't really meaningful. The ruling said Google is a monopolist and used its monopoly to maintain its monopoly. And therefore, you know, under the Sherman Act, it's liable and we have to take action. Then there was the ask from the DOJ as to what to do to break the search monopoly up. And that was to divest Chrome and share data and separate AI training

SPEAKER_01from ordinary search queries and many other things. And AI got mentioned, I think, 27 times in that. And then when Meta issued his decision on what to do to Google, which was basically nothing really, it was a bit of data sharing. Yeah, there were 72 mentions of AI in that. We had a trial about a search monopoly and we had a decision about AI. Is that even defensible on appeal? That feels like we had a completely pointless trial because it turned out that

SPEAKER_01the ruling and the remedy decision was divorced completely from the entire trial.

SPEAKER_00Yes, that is an observation I broadly share. As I said, the timing was in so far unfortunate. And I mean, the reality is during that timeframe between the liability judgment and remedy judgment, of course, there was a huge Google lobbying going on that basically run around the world and said, here we are, look, There's this huge disruption from generative AI providers, chatbots, open AI perplexity.

SPEAKER_00They are all over the shop. Never has been more competition. Therefore, you know, search in the broader sense. This is all one and the same thing. And if you now kind of make it more difficult for our loft search engine, Google, then China will lose and we are lost on a global scale. And something must have happened in that time frame that really made meta nervous. Am I now, you know,

SPEAKER_00Limiting The Abilities Of Our National Champion Google Search Too Much So That In A Few Years Time Some People May Say Look You Have Left The Field Now To The Big Chinese AI Providers That Have Taken Over And Now The World Is A Worse Place And somehow I think he was simply not certain. And he wrote that pretty clearly in his remedy judgment. What is now actually the driving force here? Will Google be able to defend its search monopoly by simply embedding features into it?

SPEAKER_00then I think the case would have been should have been rather clear we need to have strong structural remedies to make sure we fix the harm from the past or is it that if I'm now too strict with Google search and somehow other players that I like even less may step in and take over and I'm a bit scared and therefore act in a cautious manner and say well I'm rather on the conservative side here so that was

SPEAKER_00a disappointing outcome I think he was simply getting a bit afraid of his own previous analysis and was not too sure about the future of AI and its impact on search.

SPEAKER_01Can I drive down on this a little bit more? So I'm not a lawyer. I've spent a lot of time in courts and covered a lot of court cases and trials. That's not the way the courts are meant to work. I'm not sure that the judge should really be thinking about what the commercial reality of one commercial entity should be or even what the impact would be on the GDP of the United States and the country that they operate in. The job of the court is to enforce the law and there was an awful lot of talk in the trial

SPEAKER_01by Jonathan Cantor gave an incredibly impassioned talk speech about how publishers were had to be protected they were the biggest victims they were tens of billions of dollars out of pocket it had been an intentional takedown this was deliberate here are the emails internal emails I published hundreds of them you know this was about dealing with a wrong that had been done to an industry that was now having a severe bad impact on competition in the market as a whole.

SPEAKER_01That's what the trial was about. It wasn't about protecting America's best interests against China or ensuring that Google remain profitable. And so I find it indefensible for a judge to make those arguments in the final decision because that had nothing to do with it. We had a trial about somebody that admitted stealing a loaf of bread. Somebody lost out on that loaf of bread. But now we're going to let them off because they're opening a bakery down the street. That's not what the law is meant to do.

SPEAKER_00I totally agree with this analysis and I'm from a legal perspective very very disappointed with his judgment I think it's entirely inconsistent with his liability judgment and he messed up the case on the last mile for some considerations that I simply don't grasp whether it was political impact or whatever there was but certainly he got

SPEAKER_00Afraid and that is, however, never a good guidance for a judge. He should have stuck to his own analysis, should have realized, look, if I don't step in here and impose very effective remedies we will just continue seeing that kind of conduct even though it's illegal he should have done that he didn't do so hopefully and the appeal will show that this was not a good reaction to it

SPEAKER_00but I should also not forget there is a level of discretion he used it in the wrong direction for whatever reasons but unfortunately that description cannot be fully precluded.

SPEAKER_01Yeah look I've obviously not spoken to him but you know if you are sat where he was sat deciding the entire future of the internet at the era of AI that would keep you awake at four o'clock in the morning. Absolutely. What you'd probably choose to do on reflection even if you didn't want to is you'd probably come up with a ruling or a decision that enabled this to continue forward knowing full well it's going to come back to appeal and be overturned. I kind of wonder if that's kind of where it is. Let's wait for the actual appeal to happen when some of the kind

SPEAKER_01of chaos has settled a little bit and the world's a little bit more organized. But I wanted to ask you this. Let's return to the case in Munich because we talked earlier on When bad things happen to society, people first turn to fixing it themselves. But if the company is too big to fix, they can't do it. After they've done that, they normally turn to politicians and they vote and, you know, politicians fix it.

SPEAKER_01If politicians don't fix it, then you start to look at regulation and you look to experts to fix it. And if that doesn't work, you eventually end up at the court. And I've often tried to explain to people that When tech gets as big as these companies are, these are the biggest companies in human history by a considerable distance. Even if you go back to your history books and you read about the Gilded Age and you read about the railroads and you read about the Rothschilds and the JP Morgans, that was just America, right?

SPEAKER_01This is the whole world. And so in that, the way I always describe it to people is that the wave that this represents is so enormous That it rolls over people incredibly quickly. It rolls over politicians incredibly quickly. It rolls over regulators incredibly quickly. But the law is a rock. Yeah. And that's the one that's supposed to stop it. That is the one. That's the one thing that exists for exactly this reason to be the thing that stops it.

SPEAKER_01Yeah. Do you think that it's possible now for the law to stop these companies? And when I saw this happen in Munich, my immediate thought was, ah, there it is. There's a glimmer of light. There's a court that's read it, understood it and said, no, you are a publisher, clearly a publisher. By any measure, you're a publisher. In the court of public opinion, you are clearly a publisher and therefore I'm going to deal with it. Are you optimistic about that?

SPEAKER_01Is this the start of the rock wall that stops this?

SPEAKER_00I think this can be answered in a more general and in a more specific manner. Specifically on this issue, I do believe this is the beginning of a shift. I think judges will find it difficult to disagree that AI overviews are genuine content of the platform that provides the answers and that therefore liability needs to be there and you can't hide behind your platform status.

SPEAKER_00I think this will hold more generally whether law can now team these big big platforms, even though I'm a lawyer and I was active in cases and also one cases, I also observe the limits of the law. And reality is these companies are sound so big now. So Powerful politically, financially, that sometimes,

SPEAKER_00as far as I see, the law in effect does not apply to them because even if they are found to have infringed it, They find their way around it. They pay a fine, take slight changes at the fringes, but basically continue the conduct. No one follows up. Their authorities have such an information asymmetry. They don't fully grasp what's going on. The big firms are the only ones that really understand what's happening.

SPEAKER_00using that entirely to the benefit where they see resistance coming up. They tried to fix it throwing a lot of money into lobbying directly, generally through institutions, through think tanks and all the rest of it. So it's a huge I'm a huge wave that comes with those firms that tries to shape the law in their favor everywhere we are and we've seen so often in the past that there was some goodwill of legislators of authorities but then

SPEAKER_00Result of major lobbying they came to some compromises that then basically meet the entire project pointless and the outcome then changed nothing and unfortunately this is an issue that I don't see is currently fixed. And if we don't get this straight, limit the abilities of these large firms to influence how law is made, how of

SPEAKER_00authorities act and how decisions are made and implemented, we will find it very difficult to hold up the rule of law to really be that rock that stops them from creating a lot of harm to society.

SPEAKER_01It really feels like that moment, doesn't it? Because we've watched them build their businesses on old technology. Now they've got an almost infinitesimally scalable technology. And it feels like you either put the guardrails on now or this might just escape the bottle altogether.

SPEAKER_00Yes, I unfortunately fully agree. I think we still underestimate the force of this technology that is currently rolling over us in every area of life and the power that comes with it. the means to manipulate individuals, the means to shape society in a manner we may not like, the means to undermine democracy and plurality and everything we value. I think everyone really needs to wake up and realize that this is a wave

SPEAKER_00that may become too powerful for us to do something about it.

SPEAKER_01All right. Well, I've got two more questions, but the first one goes back to the German case. So can you just explain to the listeners where that case is at now? So the judgment's been made, but there's going to be an appeal. So it's not all certain. And is it going to be a precedent? I guess that's the question.

SPEAKER_00If this decision goes to the federal Supreme Court, the court will realize what we've said about the importance of general search for navigating the web does not apply to AI overviews on the contrary they will realize that AI overviews are the killer of this crucial navigation functionality of general search it's the very is the opposite of allowing free access to information. It's the very opposite of what we wanted to protect with our rulings a few years back.

SPEAKER_00So they will see that this does not apply AI overview views to AI overviews or AI mode for that matter. And therefore, I believe the chances that this will be uphold are significant.

SPEAKER_01Bravo and well said. All right. Well, my last question is this. As an Englishman speaking to a German, who's going to win the World Cup?

SPEAKER_00Sorry, that is too political. Whatever I say will be held against us. I keep it diplomatic. The best team shall win and it will fully deserve it.

SPEAKER_01beautifully oh and you I can see you in the court that was very smooth fantastic Thomas this has been hugely enlightening to me and thank you very very much indeed for investing so much of your time it's run quite long but I think it's a really really important topic so I enormously appreciate your assistance thank you very much thank you for

SPEAKER_00the invitation thank you

Open the transcript file

Hosts and Guests

Nobody is credited on this episode yet.

Credit the people on this episode

Be the first to rate this episode

More episodes

All episodes
  1. The coin flip that saved local news

    Nov 19, 202658 minAudio

    A decade ago, nobody would have bet a krone on Amedia. Norway’s biggest local news group was bleeding money. Print advertising was collapsing. Nearly one in three journalists had been let go. Its owners wanted out, and...

    Credit the people on this episode

  2. Lawsuit accuses Google of negligent design in schools

    Nov 17, 202652 minAudio

    Emma was 13 when Google’s AI first answered her. She hadn’t gone looking for it. Her school in North Carolina had blocked Gemini and ChatGPT on her Chromebook. But Google Search was still switched on. And when Google...

    Credit the people on this episode

  3. The news agency on Russia's doorstep is about to go dark

    Oct 28, 202618 minAudio

    I’m standing at a fence in a forest in south-east Finland. It is very cold, and what lies beyond the razorwire is Russia. Twenty-four hours ago I was in Paris. A late-night call from my friend Wolfgang Nedomansky...

    Credit the people on this episode

  4. Australia doubles penalties to push under-16s social ban

    Jun 28, 202628 secAudio

    The world first ban on under-16s using social media has stepped up with Australia doubling the maximum penalty for platforms. It’s been six months since, Facebook, Instagram, Kick, Reddit, Snapchat, Threads, TikTok,...

    Credit the people on this episode

  5. The Court of Karma ends Google’s $4 trillion free ride

    Jun 16, 20263 minAudio

    Stop what you’re doing and take a moment to imagine you’re the CEO of a US$4.3 trillion tech giant. Massive salary. Kudos. Keys to the city. A couple of Presidents, PMs and dictators on speed dial. A lifetime of free...

    Credit the people on this episode