E685 | This Week in European Tech with Dan, Mads & Lomax

0:00Welcome to Upside, where we look behind the headlines in the news affecting European venture. Today, it's Mads, Lomax and myself. And we're looking at Davos. We're going to drill down into what's happening for Europe. Look at the 28th regime. It is happening. Has China already won AI? A science collapse in the US. We've got a 2026 year of the IPO, not the year of the horse. SAS, is it dead already? We'll talk about that. And plenty of mega deals of the week.

0:41This show is not investment advice, and the hosts of this episode may be invested in the funds and companies featured. Gentlemen, my lovely people, how are you today? Mads, you just flown in from sunny climes. How are you feeling? Great. I just had a great week. I met lots of interesting businesses and investors in the Gulf, talked about investments, talked about our portfolio. lots of synergies yeah just lots of opportunity for the businesses in the region it's a fast growing place wonderful, wonderful, Lomax how you doing dude? yeah I'm all good yeah this week you look perplexed we've got a big storm coming through so I don't know whether the sun's out now Nazare there's going to be 100 foot waves tomorrow so I'm going to be driving up with the kids to watch which would be pretty amazing are you doing a 100 foot wave?

1:30well I'm not doing Tell me that we're going to see images of you surfing a hundred footer. Do people... Like any VC, I'll just be sit there, you know, watching on the sidelines. Yeah, telling everyone else what to do, right? Exactly, yeah, yeah. But in all seriousness, is there a size that people just don't do? No, no, people want to go as big as possible. Oh, my Lord. For sure. They want to set records. All right, well, listen, let's crack on. We've got a busy docket today. First up, something that I didn't see up at the front was the fairly recent news, but this data centre kerfuffle. So there is a£1 billion data centre that was planned for Iva, which is near London, and it's been stopped.

2:13The government has admitted that it made a mistake in granting the planning permission. So campaigners opposed on energy use and environmental impact and water, which we all understand, but it was all guns blazing. and there aren't that many of these mega projects going and we all know that we need them. I think there's maybe another four or five, including this Greystoke project. So how can they survive planning scrutiny and or are we just screwed? Mads, can you start this one off? High level, this is bad news, but there's some good news coming through, which we'll touch on at the beginning so we can end with some upside.

2:49But the bad news is the UK is under pressure for reasons we've discussed before to build out its data center capability and is majorly constrained both by energy, but also by local planning laws. So as a reminder, data centers historically have to go through the normal planning permission channel in the UK or zoning as the Americans call it. And that's done at the local level, which tends to mean that you end up with a lot of local resistance. This is where the NIMBYs come in, right? They don't want a data center in their backyard. So this is just an example that came up in the headlines today, where a£1 billion data centre being built just outside London, which had been initially approved after an appeal by central government, they'd actually found a way or came up with a way not to have a specific environmental report saying, well, it wasn't needed.

3:40But now a judge has said that, and the government have admitted that they were wrong to make that assessment, and they now need to go back and conduct the environmental assessment and it may be ultimately that this is this is approved but this is just like another spanner in the works for people trying to build out critical national infrastructure because this will be years off now right if it is approved it'd be a massive delay right yeah interestingly though i would say so this went through originally the genesis of this project is it went through local planning process right as i said it was turned down originally and then the developer appealed and it was approved it seemed erroneously now by central government was approved now very interestingly and excitingly, as of literally like this week or next week, data centers are now being designated as national significant infrastructure projects.

4:27The law is actually coming through, which means that if you are building a data center, you can forget about the local planning regime and go straight to central government for what's called a developmental consent order, a DCO, which would get you an expedited consent, would also help you do all the compulsory purchases of relevant land that you need, et cetera. It's a very strong instrument that the government has kind of put in place now for data centers. So this is really good news going forwards, which means that for people building data centers, you know, this year, next year, et cetera, they can now apply straight to central government rather than having to faff about with the local planning regime.

5:03So I think that's good news. Yeah. Does that remove the need for the environmental impact assessment? That's a good question. I don't know the answer to that. I think it will certainly make it a lot easier because a lot of this will fall at the discretion of central government. And they won't have to have all the local rules apply. However, the bats and the news. Yeah, but even DCOs, it will still be subject to judicial review and everything. Right. So you still if people want to stick their oar in, they can still appeal against these things. So it's not going to be plain saining, especially if you want to keep, you know, London is the biggest data center market in the UK.

5:38It's like 60 % of data centers. well in europe yes true but yeah obviously in the uk too you know especially if you want to keep building data centers in and around london you're going to have you know chat reviews the whole time which is now why people are starting to build them you know and there's a big big 10 billion dollar project up in in northumberland now obviously one of the big you know we have these now ai zones that have been launched following the matt clifford review last year and those are going to be located mostly in wales actually not around london which i think is sensible as long as you can get the power to them.

6:09But guys, they don't have to be around London, do they? They can be in non-NIMBY-esque areas. Well, which is why, which is why, but it's an access to power, right? And that's why I think people are looking for, thinking creatively, looking for former factories, places that actually have access to power. All the brownfields stuff. Yeah, makes sense. Makes sense. Mads, Mads, what have you got on this one? Well, I think the EAA question, the environmental impact assessment question is important because even if you have a growth zone, if you still have to go through the EAA And if that still becomes subject to consultation and objections, you're going to be into the same NIMBY theater.

6:42Yeah. So we've got another 10 years. For me, that's the essential piece. My impression was that that wasn't fully solved here, but you're probably, Lomax says there are residents, legal eagle. You're probably more up to speed on the peculiarities of the law. Even if you go through the DCO route, I'm very, I suspect that there will still be the environmental impact assessment survey, which is always going to be a problem. So it's a huge problem. It's a huge problem because it's asking the wrong question. It's designed to decide whether to permit something. But we've already decided that we should permit these data centers.

7:17We have already decided that building stuff is essential for our economic future and AI is essential for our economic future and we need to build this stuff. And so putting in place an instrument that's trying to decide whether or not we should do something we already decided to do is just slowing us down. And that is exactly why we have, you know, struggling to eke out 1 % growth instead of 3 % or 4 % growth. Now, people were very quick to attack the civil servant that made the mistake of permitting this initially. And so they're saying, look, it's a logical error. But I can see why people have done it.

7:48because the message from on high has been, we're going to root out all the blockers, all the people that are going to stand in the way. And so the person in question, man or woman, I don't know, thought they had found a way to bypass this EIA step. And it turns out there's case law that prohibits this. There was a kind of a leveling up and regeneration act in 2023 that removed the power of ministers to amend the EIA recs by a statutory instrument. So without going through a legislative process, making it harder. I mean, otherwise the government could just have tweaked the rules, but they can't do that anymore.

8:25So we've been tying ourselves, you know, in nuts over this stuff. And so, I mean, kind of all the commentary around the new fast tracking for data centers is good. I hope they're going to go the whole way. We need this stuff and fast. Yeah, and it's a reminder that, you know, while we in this podcast and people we do business with, they're all pro data centers growth, et cetera. You know, we have 3 billion, sorry, 3 gigawatts of data center capacity at the moment. We're trying to grow that to 8 to 9 to 10 by 2030. But there's a bunch of people who aren't interested in this, right? And they're more interested in the EIAs, the environmental impact assessments, et cetera.

9:02And so unfortunately, there is a little bit of a lowest common denominator that applies here in terms of the rollout. So, you know, the government are going to have to be really, really bold on pushing this through. Yeah, I think it's also on us to talk about a different narrative here because I agree with you. There will always be a small proportion that are more interested in the newts than the economic prosperity of our country. But I promise you, most of the citizens in this country will be more interested in, do we have economic growth? Can we get our deficits down? Can we get our tax rates down so we can get business going again?

9:32Can we create the jobs we need? That will be the interest of the majority of people. But the problem is we've created the legislative system where it's always, as you say, the lowest common denominator and a small minority that can block things from happening. And it's just, it's poisoned for prosperity. I've got a feeling they'll find a way of, they'll find a way of pushing this through. I think it's just, it's too big an issue now. It's too big a nut to crack. This particular data center, I think, but when you read the reports will actually end up going through. But, you know, if you lose nine to 12 months, like think about it, you have financing lined up, you have the, you know, the construct, you know, there is like so much construction going on now in data centers.

10:09If you have your project delayed by nine to 12 months, Well, what if your contractor may go and work elsewhere? You know, and all of this stuff, all the second, third, all the consequences are very, very big. But ultimately, I'm very interested to see now that we have this new DCO, like developmental consent order straight to government, central government, application process, fast track for data centers. I'm very excited and interested to see if that actually speeds things up. Just hope so. Guys, I want to talk about Davos. I was watching all the talks and all the panels and I don't normally tune in, but Mark Carney was the chap that got me to really pay attention.

10:47I absolutely loved his speech. Did you guys see it? Did you watch the recordings? I think it was by all accounts the knockout speech of Web. It was. I think it was, I mean, watching Besson and Trump and others. And there was a lot of posturing, as you always get. There was a little bit of performative kind of gamesmanship, which, of course, you're going to get because you need the click rage. But Carnage, I thought, just knocked it out of the park. There was obviously more than him going on at Davos, but I thought he was just a masterclass in statespersonship. And it was just a beautiful call to action.

11:23His point being, basically, the world is fractured. We need to accept it and move on. And he is offering a solution for the middle powers, which I thought was an interesting take. other items that caught my mind. Some obvious, some maybe not so obvious. Obviously, we need to lean into this strategic autonomy, stuff we already know. The only question here is, can Europe afford it? I saw the PWC's global chairman, Mohamed Kande, said that nearly or just over half of companies are seeing little measurable benefit from the current AI investments. So he was challenging hype versus execution. I read it as 44 % of installations are succeeding, but you can read it however you want to read it i think that's pretty good watching musk i don't think he said particularly that much i thought it was a lot of self-promotion obviously talking about ai and robotics how how his uh robots are going to make work obsolete which i think is utter nonsense i often think the chap forgets how to be a human being will we see tesla's world of abundance as he's pushing lots of talk of human ai human-centered approach to ai i have no idea what that means and the IMF is warning of any kind of correction in AI, causing much more of a rift.

12:33So where should we start here? Mads, do you want to kick this one off? You're right. So many treasures to gems to pick out. I mean, Carney clearly standout speech. And I think the thing he said was, look, the old order is not coming back and nostalgia is not a strategy. And I think we can all agree with him on that. What was interesting is he was talking about middle powers and he sort of said, look, middle powers shouldn't be negotiating bilaterally because, you know, I think what he's talking about there is if the European countries and Canada and others kind of middle sized countries, they start doing individual deals with China and with the US, we're going to be picked off one by one.

13:14And that's going to be that's going to be the worst possible outcome for us. Which is kind of funny because actually that's in the context of the UK. That's kind of been Starmer's strategy, you know? Like he's been leaning into a kind of very bilateral approach, right? And it works right up until interests are no longer aligned. And as you saw, Trump had absolutely no prevarication about threatening to slap tariffs on the UK if we are saying that we're going to stand up for the sovereignty of other European countries. I mean, it's the strategy of empire since the dawn of time, right? It's divide and conquer.

13:48Split them up and have them pick off one by one. Europeans are getting a taste of their own medicine now, no? Oh, absolutely. And now I think if you think one step further, I would say there was this idea that being a middle power that could punch above its weight was once seen as a viable strategy. That idea is now dead completely. I mean, you have to unite and form stronger blocks. There is no space for middle powers in this because you're just going to get steamrolled. So unless you can work together and collaborate, you're toast. And I think that's what we've seen without the shadow of a doubt now.

14:27And this is a new world, right? I mean, we used to be able to play the middle power game because interests were sufficiently aligned with the U.S. that you could sort of, you know, you could be under the protective blanket of U.S. military supremacy and sort of, you know, find a way to navigate. But, you know, if that's threatened to be taken away at any moment's notice, you're left naked. And then you can see what happens. Many, many other things to pick up on about kind of Canada's relationship to Europe. If you want to go there, there's Jensen Wang on robotics, which I thought was really interesting.

14:59There's AI and jobs. There's developers and what's going to happen. So Dan Lomax, what's top of the staff for you? What do you think, Mads? Jensen was blowing a bit of smoke, blowing, you know, like as in... I read him like that as well. I read Talking Your Own Book, Talking Your Own Game, waxing, waxing. No, but also just like being in Europe and also like trying to praise, you know, a bunch of his customers in Europe around the robotics opportunity in Europe, no? I just didn't, I didn't find any meat in it. Listen, I thought what's interesting is you're sort of saying, look, you know, forget competing on foundational models, but Europe's opportunity is physical AI.

15:36You've got the engineering heritage. You've got the legacy of making, you know, the best engineering and machinery and technology, kind of hard physical technology in the world. Lean into it. Become the best at that. Now, the problem is we have been selling some of our crown jewels. We sold KUKA to Medea in China. We sold ABB Robotics to SoftBank. We've been selling out all these robotics companies. The ABB deal was at the end of last year as well. It's a recent thing. It's very, very recent. And so, you know, 1X from Norway, they moved their headquarter and actually all of their operation from Norway to Palo Alto.

16:07So it's clear that we haven't worked out fully how to lean into this opportunity yet. But I think he's telling us what we need to do. We need to lean into the stuff we're good at. We need to put the AI layer on top of our manufacturing capability. Well, it's following the Chinese way of doing it, isn't it, Mads? This is what the Chinese have been doing for a little period of time now. Yeah, I think they come from a very different perspective. And the question is, you know, what will be required for us to catch up and can we do it? What they needed to do, they were behind in engineering and technology.

16:38And they've proved by really focusing on those things, they could catch up. I think we already had so much of the tech, so much of the engineering capability. our problems are not technological, they're political. The problem we have is all the fragmentation. And capital, right? The capital is here, Dan. The money is here. I know, but it's just not going in the right direction. Which, again, is a political problem. So it all comes down to how we have organized ourselves politically. And I think Jensen, he talks about it as an engineer. He's like, just focus on the stuff you're good at and things will work well.

17:10I think the challenge we really have to grasp is how do we organize ourselves politically so we can unlock the engineering capability, the commercial capability, the capital, as you say, Dan, we need. That's purely a political question. Is there will? The fear one can have is things are not yet bad enough for us to take some of the difficult decisions that are needed. But at some point, I mean, hopefully we will do it because we can't continue working or not working the way we are. Two challenges to this notion of selling the crown jewels when it comes to robotics, Mads. It's like, well, firstly, those companies still are in Europe, right?

17:46They're still employing people, building, manufacturing here. I mean, I guess the risk is till the IP gets stripped out, but they're still on the ground in Europe. Number two is, should we be indexing on these companies, like, you know, subsidiaries of ABB Robotics? You mentioned One X out of Norway, and it's a shame they've moved to the US, but there are plenty of amazing robotics companies coming through now that you and I know well, and and some of them we don't know, across Europe. And aren't they the kind of future champions of this category? No? If we can get them the capital they need, if we can make it easy enough to do business, if we can allow their business partners to build the data centers with the AI capabilities they need to train the models, if we can get out of our own way, then absolutely yes.

18:29I agree with you. I believe we have the impetus, Mads. I do believe we have the impetus and have been given permission. I do believe there might be some inertia, but I do believe we will do it. Here is my bear case, Dan, just to share it with you and bring it out there. Until we have a European leader with a democratic mandate, we will not be able to cut the gourd knot of this. Are you talking about the United States of Europe? Are you talking about... The leader of the Commission or the European government has to have a democratic mandate. What happens every time you see somebody in Brussels say, listen, we've got to federalize here.

19:04We've got to take a joined up approach. It fragments, it runs into local opposition, it's stifled. We're going to pick up on EU Inc. later. I promise you that is going to face a world of hurt when you go to local implementation across the region. Mads is clearly a subscriber to basically the big band theory of history, which is that big people at various times in history, I think there's a very Victorian approach to this, but some historians really subscribe to this and some don't, that it requires like big men, historically men, but it could be women too, to drive big things at big moments in time.

19:42And if you look through history, that's what's happened. Actually, Trump has kind of brought back this kind of like people who believe more and subscribe to this theory of history, right? So Mad seems to be clearly subscribing to that. And I don't disagree, by the way. I mean, you know. Well, you know, from memory, I think both Elizabeth and Victoria were women. And obviously - No, sorry, I'm just using - I could use the word big person, but historically has mainly been named. To be honest, I think sometimes it's more about how you organize yourself. Of course, you have to have the right leader and the right leadership, but the institutions have to be set up to enable those leaders to be successful.

20:19We can think whatever we want about Van der Leyen, but she does not have the institutional setup to succeed with the reforms she's trying to do. Nor the personality, it turns out. I've been reading the 21 Lessons for the 21st Century, the Yuval Noah Harari book, and it does talk very much that kind of game and how Putin's and the Trump's and the big people have triumphed over history. But one thing I did want to touch on before we move on from Davos is this human-centered AI, this job destruction, this employment issue. And I saw Demis talking about it and Dario talking about it. But Mads, what would you pick up on on the AI and jobs piece?

21:05I mean, the headline that is having everybody running scared right now is it looks like entry-level job postings are down since Jan 23. and so people are saying, look, it's... Yeah, entry level. Sorry, entry level. Young people, exactly. It's kind of the graduate jobs, people out of uni can't conquer jobs. And the big question is, is that cost by AI? It seems that there is suspicious correlation with the end of SERP and the increase of interest rates that have made companies slam on the brakes of some of this hiring. We can all see the impact of AI and how much more efficient it's making us. Now, I personally happen to think that as we become more efficient, we're going to want to do more stuff.

21:45It's not like we're running out of problems to solve in this world, but businesses will be organizing themselves differently and that can have implications for how they hire and how they structure themselves. And yes, there could be an implication on the job market. I don't believe in mass unemployment and I do believe there will be shuffling as we've seen in engineering and coding, but I don't believe we're going to see this kind of massive dearth of jobs. And looking, there was a very interesting article in the FT taking some of those reports to pieces a bit and connecting that Zerpi period of time with the launch of ChatGPT and trying to make sense of a bigger picture rather than just saying AI is killing entry-level jobs.

22:26But Lomax, what would your take be? No, nothing really. I think it's more related to the macro economy than to AI specifically in the context of the entry-level jobs you know LLMs had really had barely scratched the surface when when these numbers come from in 2023 so I can't imagine law firms accountancy firms any other white colored jobs making those decisions in that time based on what was happening in in AI right yeah it's kind of now going forwards that you might expect to see that but I think it was more macroeconomy driven let's let's get into another announcement at Davos and has been talked about we've talked about a lot and it's been going around the social web for what feels like millennia, but only probably in the last 18 months.

23:10The 28th regime. So Von Alain announcing the 28th regime is game on. It's a pan-European legal framework for startups. For those that really genuinely don't know, but I'm sure you do. It's go time. Formal proposal going forward in March. Should we start sobbing for the notaries just yet, Lomax? No, I don't think the fat lady has definitely not sung for the notaries yet and may not do so, despite the optimism that there is around this. But it's great, great momentum around EU Inc. As a reminder, there are 27 member states in the EU and this is called the 28th regime as a result because it's effectively a new legal framework just for corporation, company law.

23:52This is to provide a simple legal entity which startups mainly can use, cheap to set up, easy to set up. The proposal is you can set it up and register it in 48 hours. There's no minimum capital requirement. You know, in many European countries, you have to put in at least 15 grand into a company just to get it started, right? You know, much quicker, simpler, easier legal entity, which will kind of reduce friction and time and cost for entrepreneurs. So I think that's... And investors, of course. And then around that, there are other initiatives, which is the ESOP, the Employee Share Option Plan, you know, which is both...

24:31And ESOP is a sort of conglomeration of law and tax, right? And the problem is we have those conglomerations across 28, 27 different countries at the moment, right? So there's a simplified ESOP proposal as part of the EU Inc. proposals. The slightly complicating thing is that for ESOP's tax element is still a local competence. And the EU does not have jurisdiction over tax. But what we're really trying to say is, as part of the EU Inc, is let's make ESOPs simple, clear and defer tax to when people actually get paid rather than giving them these dry tax charges on exercise of options, etc. So there's a drive towards simplicity and also fairness.

25:15And I think that's, you know, that's part of the proposals. Initial vote put through this week, it now needs to go through the legislative process, which could be 12 to 18 months. this is where the pushback will come. So it's not unanimous as to this that this is going to happen. I mean, the vote this week was very was very favorable. I think one of the biggest questions now is more a process point of EU law is will it become either a regulation or a directive? Our reminder is a regulation will be binding directly on all member states. And a directive is a directive which leaves to the member state the discretion to implement it in their local laws.