E409 | This Week in European Tech with Dan Bowyer, Mads Jensen, Lomax Ward and Andrew J. Scott

30:19I might calculate that I need 100 million pounds in my plan 20 years from now to have enough money to pay out pensions. And I might have 50 million today. And I'm going to have to make some assumptions around how that money is going to grow over time to be worth enough that it can cover pensions. And so the question is, when is there enough? So it doesn't feel like it's very thought through to me. And this is not kind of, Andrew, to your point from earlier, I think it's very easy to become party political on this. I saw a headline in The Telegraph the other day saying, well, the biggest threat to growth is Rachel Reeves.

30:59But it's not exactly as if the other team was doing very well either. If you look at the UK's economic performance over the last five, even 10 years, it's been pretty abysmal. We even had a prime minister who I think when asked about what business thought said something like F-U-C-K, business. So I actually think on both sides of the aisle, it's been pretty abysmal. And I just think we should expect a lot more, starting with a fundamental understanding just of how business works from our politicians. Not one party or the other, but frankly, all of them. Well, none of them worked in business to your point, Andrew.

31:32None of the Labour cabinet have ever worked in business. So it might be a bit of a tough conversation. The other thing that I kind of was, when I was reading through all the reports, was Reeves was talking about this pension surplus going into riskier assets. And it made me think what they deem riskier assets. Because I always wonder how they perceive and how they understand how innovation works and where that newness comes from. Yeah, but a simple way to think about it, Dan, is if you have a company that runs a pension plan, then that company will be funding that plan. And effectively, what it's saying is the company can then take some of the capital back.

32:12And presumably, that is then to invest in the company. Right? So I run a manufacturing business. I put 50 million into the pension plan. I can now see there's a surplus. I take 5 million back out that I can reinvest in my business to help make it grow. Oh, so it's direct. Yeah, yeah. So I think that the principle of it makes sense at a very high level. It's just that because of the way we have fiduciary obligations, because of the way trustees operate, the detail is really complicated here. There's going to be a small trickle through line. I think like, you know, Small fry. I agree. Reid's announcement was an aim to steal or start some momentum with some PR-grabbing headlines.

32:53As Andrew can be pointing, Heathrow isn't going to be on mind for another 40, 15, maybe 20 years, right? The pensions, high-level numbers are massive, as Mads said. It's amazing that the defined benefit schemes, which we're getting rid of in this country, right? Because the idea of promising somebody for life a fixed amount of money on their final salary, people realized was an insane concept, right? Especially seeing we're now living into 100s, it's even more insane. Exactly. So they're being sunsetted, right? That they exist. And they historically, you know, when I graduated, I remember, you know, working a little bit on this.

33:28These things were all in deficit and it was a major problem. Now we have to benefit after the low interest rate environment that they're in surplus. And, you know, there is quite big surplus. we're talking about and the defined benefit, which is roughly half the industry, at least 100 billion. But no one's going to put more than like one or 2 % of that into venture. And these people don't, I mean, not these people, but these like very smart, I mean, potentially smart, but very risk immersed and trustees who have all these obligations. I mean, we're in the weeds here. We, you know, like for most people, like venture is incredibly, they think normal private equity is risky.

34:05We're like off the spectrum when it comes to risk. So I think it's great. I'm not going to say no to the money. But what will happen, by the way, is that the money will come and it will be in the kind of low single digit billions over a few years. And it will be allocated to, quite frankly, guys, the most boring risk averse like allocators of venture in Europe. Right. So it's probably not even going to be that exciting. I'm not going to name any names, but the kind of more risk-averse kind of funds that I think exist to gather assets under management rather than performance fee. Name some names, Lomax.

34:42Get it out. I'm not naming names. I think your point about where the money goes is really interesting, though. And I happen to work like you in the riskiest stuff in this D-Tech, the game-changing technology. But of course, there's a place and a need to fund the more established, predictable markets like fintech in Europe and London is a good example. You mentioned Swan earlier. There are probably, what, over 100 banks in Europe with assets under management of over 100 billion. I think the biggest is HSBC with over 2 trillion in assets. So there's a huge amount of opportunity for fintech challenges and platforms and banks, new banks, and a clear path to exit and for M &A within Europe.

35:30So that's a good thing. The problem is we've just got to also fund the really risky stuff unless you want to be owned by America with AI and quantum and everything else. Listen, I want to get onto DeepSeek because that's a very natural segue into what's Europe's position now that DeepSeek has ruffled some feathers. Yeah, I think there's maybe another point to touch on. We will cover and we must cover DeepSeek. But just while we're on the UK government, to Andrew's point, there is a surprising lack of creativity in the thinking we're seeing. we have the most powerful technological revolution in our hands right now with ai and we are looking at kind of these massive areas that absorb funds right like the nhs there's massive black hole we keep putting more and more money into and then there is this whole notion that we almost can't do any innovation because we can't afford it when really the thing we should be doing is being extremely aggressive about automating and applying AI.

36:38You can't afford not to. You literally can't afford not to because we cannot even afford to run the NHS to the service levels that people have come to expect and that you've had in the past because we are going broke. So it seems to me that the thing one should be doing if one is in government right now is to be super aggressive about really pushing the AI agenda to the frontline services in health, in education, in security, in everywhere where you can to create economic opportunity. And that will then trickle through to innovation everywhere. They are talking about it. I mean, this is, I think, maybe to Andrew's kind of feeling gaslit point.

37:19I've never seen a government release so many PR articles about investment, growth, innovation, but the challenge is being brave enough to actually do it. Whereas the policy, all the departments need to understand that growth is the number one goal. And underneath that, there should be no more than a handful of initiatives, which we believe will ensure we achieve growth, which of AI, which should be one, clearly. a couple of things to end on like don't forget we have a later government in the uk remember where the later government came from like are those the best people to be stewards of of growth in the uk and i'm not going to you know we start with them we start with them unless you want to start the revolution though max i mean that will be behind you in the street and the second the second thing i want to say because we i mean we sort of touched on the beginning of that this silicon valley thing right so talk about this like repreating silicon valley you know first of all or have any of you driven from Oxford to Cambridge?

38:19You're based on driving across fields. Like it's one of actually the hardest journeys going from East to West in the UK is very virtually impossible. But you do not recreate Silicon Valley with buildings, with roads, with houses. I'm sorry. Like Silicon Valley is a, you know, I was thinking about it earlier. It's a certain like savoir-faire. It's an esprit de corps. It's like a, it's a movement. It's not buildings. It's not infrastructure. And how do you kind of foster that? Forget what Rachel Reeves said. It's not with airports. It's actually with immigration. It's with softer things, to be honest.

38:57It's with smart skills-based immigration. It's with fostering talent. It's deregulation. It's encouraging success stories and actually making sure that, you know, the companies that do start to break out have the ability and maybe is funding them or if they're good enough, they should. they should be able to grow themselves. So the notion that Rachel Reid is going to create a silicon bound, that is for the birds. That is really for the birds. When I did my first company, because I grew up in Cambridge, it was Silicon Fend that used to go, we're going to build Silicon Fend. And that's still being banded around.

39:33But as I said, I've whinged about the train line from Cambridge to London many a time, including on this podcast. And you can't even hold a telephone call. So let's get the, you know, you need the basics. I'm with you, it's a mindset, But if you don't have the basic infrastructure, you're fucked. Get the basics in. I want to, this isn't a European story, but it does have deep-reaching European implications. We cannot, this week, not talk about DeepSeek. So I don't even know where to start. I've been reading and... I can start. No, Andrew, crack on. Go for it. Go for it. Go for it. I think it's a storm in a teacup.

40:08Oh, okay. Claiming IP infringement when your entire company is built on infringing everybody else's IP is absolutely big tech. So I just view this as healthy competition. I think this is classic capitalism. Why are they complaining? Just get on with building better tech, which works cheaper. They should buy my company for 10 billion and then they'll be well ahead for the next 10 years. it has uh it has been a fascinating week but mads i mean you were talking about this back in december and we've had conversations all throughout january about these incredible models and what they're doing and we can get into did they steal you know what is distillation and some of the mechanics of it but you were looking at this stuff back in december give us give us a little bit of a run through yeah so december 26th when you know most of us were busy kind of tucking into turkey and other Christmas delicacies, the DeepSeek team released version three of their model, which really was the big innovation.

41:10I mean, that was sort of the groundbreaking movement. And in the tech community, folks were looking at it and saying, well, this is actually really, really cool. It's really impressive. It's really performant. It was completely open source. The white paper that accompanied the model was very impressively written. And there were a number of innovations that I think genuinely put the Chinese model developers on the map. So for the first time, you had a team that wasn't a US team that had been able to shift the curve of efficiency in developing new models. Many folks will be familiar with the term scaling laws, sort of this notion we have that to make models better, we can make models better if we put more GPUs, more chips and more data behind them.

41:57And the scaling laws would say, Well, every time you put a factor of 10 more of GPUs and data behind a model, you can make the model, let's say, 20 or 30 percent better. And so they've been pretty immutable over the last couple of years. But what the DeepSeq team managed to do was to create some innovation in the model architecture itself that meant they were able to make the models better without just putting a lot more chips and a lot more data behind them. And that was terrific. Mads, quick question on that. Quick question on that. Is this because they were restricted on chips? They were using, what was the H100s and the models before?

42:39Or was this, do you think this was a result of the Americans stopping chip sales to China? Maybe. I think it's possibly a little too early to say that there is a real impact of that yet. It's possible. I mean, we always say necessity is the mother of invention. So whether it was because they just didn't have access to the chips or possibly more likely, given what we can see, is they have more chips than they've used, but they took the route of trying to be more efficient for other reasons. They were very successful. And that's the first time that's happened. And that has created a level of maybe frustration or upsetness in the West.

43:25It's just that we didn't pay attention at the time. And then they released R1 of the model last week and people started to pay attention. Interestingly, that V3 probably was the bigger innovation, but it just took us a bit of time. So some sour grapes now. There's been all this, as Andrew said, all this blame. People are saying they cheated. They had more GPUs. They were using many more GPUs than they've been saying. They were stealing IP. Sam Altman has been saying, although, yes, as you say, he's been pretty good at borrowing IP, I think he calls it. But Mads, it's fairly obvious they've been training their model.

44:04and it looks like they've been training their model on Sonnet and on O1 from OpenAI. So, and that's where this whole notion of distillation comes from. And that's not, I mean, that does make you incredibly more efficient. Well, okay. Everybody does that. Okay. Okay. I mean, carry on because I also want to get on to maybe the TikTok implications and the broader European implications, but carry on for now. Great. So what's interesting is that there are other really strong models coming out of China. So Alibaba has Gwen, ByteDance, they've got the Duo Bao, and they're both really, really performant and great models.

44:39And it just sort of shows that the whole notion of us trying to keep AI in the West is probably not working. Now that said, Amodei from Anthropic has come out strongly in favor of continuing the export restrictions. He says that we expect that the scaling loss will continue. and although there is innovation at the architectural level, you're still going to benefit from scale. And the chips that the DeepSeek team have been using are chips they could buy because they bought before these restrictions came into force. But the expectation is that from Amodei's point that he is really one of the thought leaders in this space is that scaling loss will continue to apply.

45:23And so he's arguing, look, we should continue to have these export restrictions. Just before we move on, there's one other point. This may be one for you, Dan, because as many of the NASDAQ, many of the Max 7 stocks were tanking last week, and especially companies like NVIDIA and Broadcom were feeling the heat. One company went up, and that was Apple. And at first, people were speculating that it's because Apple is so irrelevant to AI that anybody who is a serious AI investor just didn't care about them. I didn't care to short the stock. But there is another notion, which is that the commoditization of the models should all else equal move power to people that may not have model knowledge, but have hardware and data.

46:16And of course, that massively favors Apple. So I know you've been sort of iterating a couple of times. Or should we do? It's a random approach for Dan, is it? You know, 2025 is the year of Apple and AI. But there's actually some sense that it may come good. Yeah. So for those that haven't heard me rattle on about this incessantly, I do believe that regardless of the reality and whether they faked it or stole this or did that, it does lean into this. If you have that network and access, the leverage that you can create from these lower cost models and you can you can once you have that once you own your own network or apple has this massive network of users and reach you can really lever it so i think i i yeah on the consumer side i think that's going to be a massive massive play for them now lomax you you've got some notes on what's been happening with deep seek what what what have we missed where are the gaps no no no i i think not well i mean everyone's a deep seek expert these days if you open linkedin so i've I'll be careful about what I say rather than not wishing to contribute too much to the noise.

47:22But there's maybe three observations I would say. Madge referred to Dario Amadei and his opinions on the enhanced export controls. That's the CEO of Anthropik. He put out a really good piece, I think, today or yesterday. Also explaining how when you actually break it down, the costs, order of magnitude, is actually not way off what one might expect. so it's a really interesting piece that I would recommend people to dig out and have a look at and then three observations one is you know I think doing a truck yeah doing a troll of social media to get um to get a sentiment here our friend Nathan at Airstreet put out an interesting interesting comment um saying you know we have a a nation um tight uh with with um with great talent and with limited resources and budget or limited access, China, why have Europe not done the same thing?

48:19You know, I mean, obviously, we have Mistral, but I mean, everyone's complaining about us not having the tens of billions going into AI. But where's Europe's defense on this? I mean, clearly, we have an amazing... The fact that it was a Chinese outfit and not Mistral that did this, I think... It's a bit depressing, isn't it? I know. I know, I know. Ex on our face. Yeah. Yeah. And the founder, he used the same methodology to build his hedge fund. And they've only been going just over a year. So it's quite a crazy story if you dig behind it. Andrew, have you got any thoughts on this? Did you want to dig into anything?

48:54Oh, can I finish my point, sir? Is that all right? Oh, yeah, go ahead. Sorry, yeah. Do it. No, I think... Do it. Do it. I mean, there have been a lot of people who are not only experts on DeepSeat, but they're also experts on this concept of Jevons paradox these days, right so i think you know we've talked about how we've talked about nvidia being a big sufferer here in terms of the dent in its market cap being you know being the um you know being the measly 600 billion i think was yeah but you know ultimately i think if you step back from this i think it's ultimately you know ultimately a good thing in terms of ai like more and more smart people working on it the us have now just had a you know firecracker put up you know put up the backsides um trump will probably be picking up the phone to these guys well you know i thought you promised me dominance now you know make sure you freaking deliver yeah tariff city i imagine some some kind of yeah no i just think that you know in terms of it depends what you think about the ai race by the way like you may you may not think it's a good thing but if you think it's a good thing in that you know tech can change the world etc etc then you know things have just got even more supercharged ultimately and to go about the jevin's paradox the point was is you know the the Nvidia stock has taken a pummeling because people think that this will massively reduce their ability to ship products in the future.

50:12You have to explain what Jevons paradox is now. I will now, effectively now saying that actually, if you believe the efficiencies of the Chinese model, this will actually enhance the propagation of AI, the proliferance, we will have more and more use of AI, which actually will in turn be good for NVIDIA rather than bad for NVIDIA. Right? Because it came from the beginning of the industrial revolution, if I remember correctly, it's about coal, wasn't it? It was the fact that as machines became more efficient and they needed less coal, they sold more coal. That was the paradox, I think, originally.

50:51So, yeah, exactly. So, and you can apply that to AI. I mean, you believe in it or you don't, but I think it's certainly a strong argument. Well, the efficiencies are massive. I was looking at some of the costs i mean it was i think it was four dollars forty for a million tokens on on on chat gpt and it was like 10 cents on on deep sea car one so it was it was it was a and they were saying that what cost five billion cost them five million dollars i mean the the numbers weren't even like a 10 20 say where they were yeah i wouldn't i said read you could read dario amadei's piece on that you know i mean yeah you know he's he's got some bias but i would read it it's very interesting yeah listen it's a nice poke in the eye to trump isn't it so it's a lovely after all the tiktok debacle and this is one other thing i want to just quickly frame i mean we are effectively giving data to the chinese government i don't know what the reality is of that but if there was a tiktok challenge there is definitely a deep seek challenge sure sure you're going to run it locally if you want to get geeky and then you don't know it's open source and that's all wonderful stuff but it just feels like that you don't really know how these things work i'm not even sure the developers know how it works.

51:58But Andrew, I've monopolized too much. Any closing thoughts on DeepSeek? I just think we're so early in this journey with AI and we forget that sometimes. So there are going to be lots of plot twists and turns and there are going to be some impressive technical step changes that seemingly come out of nowhere, pretty much like OpenAI came out of nowhere but actually what started in 2015 was it the overnight success that took 10 years to get there that one yeah so i think we're going to see a lot more of that so it's going to be an exciting few years that is a very good point and actually again you know reflecting on my social media meanderings this week saranga the partner at um boulderton put out a post by this basically saying just let's not forget here guys we are we're still very very very very early on the fronty here like with no one really knows anything we're all still learning there's let's be careful um but let's learn and explore let's also acknowledge that tons of money is going to get lost in a lot of these projects right like but that's fine that's part of the that's part of the model in the industry um and i think that's that's something we sometimes forget when my dad is you know eat whatsapping me about deep seek which is quite frankly astonishing it reminds me when my mother-in-law was asking me about bitcoin a few years ago yes in the last run it's like my mom in her 80s told me to invest yeah always yeah always a time to always time to always a time to head in the other direction but it's like let's not you know we're playing around at the absolute frontier here so yeah as andrew says we just don't know a lot it's funny to think that we haven't really hit the application layers yet i mean we're talking about the foundational models We haven't really seen, I mean, 25 was supposed to be the Eurovegenic workflow, which I still passionately believe it will be, but we haven't even really applied a lot of these incredible technologies.

53:54and very quickly I want to take one step back because Mads, we were talking about was it chatGPTGov? There's a government edition of chatGPT which I was always wondering does it seem like it's doing anything out of the ordinary that you wouldn't see in the enterprise or in standard businesses it's being used at the Air Force Research Lab Los Alamos Lab, Minnesota and Pennsylvania state agencies and OpenAI launched this chatGPTGov and I'm wondering, will there be a UK version? Can you imagine the civil service using it? Is this Skynet version 0.1? Yeah, Miles Dyson's got his finger on the trigger.

54:39So I love that. I mean, there's obviously so much more to dig into with DeepSeek, but very quickly to round off. Round of the week, we've got 11 labs. Now, they've done, 11 labs are raising$250 million at$3 billion, and they're not even three years old. A lovely European AI voice generator. Now, Lomax, I think you threw this in the docket. Anything that you wanted to add to a nice big juicy round for them? No, actually, we've had now three big mega rounds, right, this year already. It's great to see these mega rounds happening in Europe. It's great to see them happening in very short timelines.

55:17As you said, this company is three years old. It's great to see employees or former employees of big tech companies in Europe, like Palantir and Google, setting up and founding these companies in Europe. It's amazing to see the momentum that they've got. I think it's just great to have it. And actually, a couple of the seed funds that got in right at the beginning, we know really, really well. And the returns for them are going to be absolutely like career making. So I think, you know, we talk about, you know, new managers coming through backing, you know, backing entrepreneurs. And, you know, with deals like this, they're probably returning many, many, you know, just a step back here.

56:00Right. You know, if you're running a, I don't know, 50 mil seed fund and you got in early, you know, if you're holding, you know, 10 percent of this company or 5 percent of this company. I mean, you're already looking at a kind of three, four, five X fund. Right. So, yeah, that's very, very. seriously i mean you know hats about hats off to the founders but hats off to you know the guys who who backed them early and that's what this industry needs it reminds me that you know the u.s venture capital industry has got you know 50 60 70 years of like cycles and returns and we are still very early in europe and you know this stuff will just does take time but you know we've seen what neco 250 last week you know these are big rounds now and i'm actually having i've been telling you guys that i'm a bit like i'm very down on european risk capital but you know what i mean actually there's maybe maybe i'm wrong good for them and it comes back to the arm the original arm story those 18 entrepreneurs i mean arm was set up in the 90s i think it might even be 1990 when arm started so it takes time for this recycling and this connection not hopefully that long but it's great that we're starting to get this full stack happening in europe right lovely p oh um andrew alice and bob what's happening there so uh european quantum computing startup uh they create they have a a type of uh allegedly a superconducting uh circuit that by design is like more resilient to noise um not many people are focused on this particular type of quantum computing technology i think um aws are interested in as well series b round led by um axa venture partners great to see them doing this sort of high-risk stuff and then future french champions which i think is an amalgamation of like the catar and bpi france both sovereign wealth funds so you know classic again where's the money coming from oh look it's the government um but just great that the round happened um we've got a competing company in our portfolio universal quantum building quantum computers but i just think that it's fantastic to see quantum getting funded actually in a way that perhaps ai hasn't been historically in europe i've seen a lot of quantum news this week there's been a lot of quantum noise and and if you headline so i think maybe ai is taking taking all the heat well guys quantum quantum stocks are up you know literally you know um and actually um quantinium which is a actually former uk company called cambridge um cambridge quantum computing which merged with honeywell um so it's half uk i would say half european is slated for an IPO this year.

58:34And people have been talking about, yeah, their last private market valuation was 5 billion. And people have been talking about 10, 15, 20 billion at IDO. So that's a pretty big outcome. And the actual listed quantum tech stops, you've got Rigetti at 3 billion. You've got INQ, I think now at what, 5, 6 billion. I mean, you've got D-Wave at 2, 3 billion. By the way, D-Wave and Rigetti were both like bankrupt at some point. So quantum's going through a bit of, it's got a bit of a shot in the arm at the moment. so long may that continue. You know, a bit like Tesla's self-driving is in five years, and there's that kind of, at some point, quantum, when we say, oh, we're about five years out, somebody will be true.

59:13I look forward to that day. And you've got, you know, UK and Europe have got some great stories in quantum, and actually CyQuantum, which is probably one of the most well-funded quantum companies in the world, you know, came out of the University of Bristol. I mean, it's basically become a US company now, sadly, and it probably is an example of where actually they had to move to the US to raise the money but you know the IP and the tech and the talent came out of both Australia and the University of Bristol so I think you know and there's some great you know quantum startups in the UK that are really showing some really really compelling data so I think that we can do a session on quantum I think in the future I think it's very exciting.

59:51Before we sign off I wanted to throw something out there actually did anyone see this Gen Z survey come out 3 ,000 Gen Z's and that's an age up to 27 years old so they're not all kids said that 47 % of them want Britain to be a dictatorship because politics doesn't work anymore yeah this is a very good point by the way wasn't that men fascinating Was that men or was that Gen Z? Was it just men? Gen Z. That's Gen Z. Because there's this like, if you look at the data for young men in particular, who's felt like very pushed out by the kind of, you know, the woke left. They are big backers of Trump, big backers of the strongmen.

1:00:43And guys, one thing we didn't talk about was Elon Musk on stage at the AFD rally this week. I mean, I think that is mental. absolutely i mean afd are now polling at 20 so they're second and by the way it's the german elections at the end of february so it's around the corner you know afd as a reminder you'll come to the deutschland the the sort of far right um anti-immigrant um german party is now second in the polls but you had elon addressing their rally um you know absolutely pumping it i'm i'm very I'm very worried. Actually I got this wrong it's worse it's worse it's 52 percent think that the UK should be a dictatorship 47 percent agree with the statement that the entire way of our societies organized must be radically changed through revolution.

1:01:34Yeah they've been listening to too much Andrew Tate, Jordan Peterson and these people are I'm afraid it's very scary and it's very relevant to us because um this is the way this is the future in in defense of current in defense of kermit the frog i'm not sure it's fair to put uh jordan peterson and uh andrew tate in the same bracket actually it's all i don't know man but peterson peterson is off the ranch he has definitely left the ranch he left 100 i'd say new new wave peterson is off the is off the reservation yeah i was a massive fan i was i was a massive fan but this does remind me i i'm a big fan of him I'm Scott Galloway, and it does remind me of he, a number of years ago, was extremely anxious around the number of lonely young men there were in society who are obviously being drawn and dragged into Trump's fears and Andrew Tate's fears because there has been almost a disenfranchisement of young men.

1:02:31and this is possibly not a topic we should ever dig into, but I think there's so many conflating challenges here. Anywho, we are - Talk about Peterson going off reservation. I wouldn't end on this. And actually, I think this is something I've been thinking about Elon recently, which is like, Elon is a phenomenal entrepreneur. And I take my hat off to the guy. So many, many things. But watching him of late with his dabblings in politics, I start to worry about Elon, to be honest. and watching him on stage in Germany, I got very scared. I'm very scared about it. And there's sort of lack of detail and it's very...

1:03:12I find it very discomforting. Well, as Eddie Izzard once said, mass murderers come from the places you least expect them. So let's hope he never turns the corner. Oh, Jesus. We were doing so well at being optimistic. We were doing so well at being positive we've fallen into that into that it's your fault you invited me on the show i did i did that was what do you expect what do you listen let's end on a positive note andrew what are you looking forward to this week what's uh what's happening for you i have for the rest of the week very few meetings i've intentionally head down some deep work very excited by that any themes or topics you want to share just uh change the world is that okay small just a small little notion lo max what's on your agenda this week the thing i'm most excited about just just closed a investment in a UPA defense company.

1:04:00So I'm starting to roll our sleeves up with the founder now and make plans and help where we can. So that's the most exciting thing, I'd say. Nice. Mads, what's cooking for you? Oh, there's so much. We've got two amazing founders that have just closed great rounds. One's just closed and the other one has just signed term sheet, but it's worked towards a closing. We've got another one we're working on, a new investment. We're making a fascinating AI company that we hopefully can talk about in a few weeks. So, yeah, so much good stuff happening. I just loving the founders we work with. Easy to criticize and be down on stuff, but there's also so much optimism and so much excitement around, and I think that's just a great thing to work with every day.

1:04:42Yeah, no, I think that's the joy of what we do is that we get to see the innovation at the coalface. Don't forget, Dan, we've got to be a bit on brand as Europeans. We can't be all euphoric and enthusiasm. and you know we've got to be a little bit miseries we've got only a little bit you know but otherwise people will be are these people even in europe you know yeah these are these americans talking funny i've got um we've got an investor day this week so we're showcasing six of our teams to to some of the investors in our in our last fund which i think is going to be a lovely lovely afternoon looking forward to that and i'm also doing a talk at london stock exchange about investing in ai and the future of ai so i'm very much looking forward to that on thursday but gentlemen it is always a gift and a pleasure i will catch up with you next week have bloody good ones thank you all have a good one catch you then bye bye tear down this wall it's more than just an ally this this is a union of values let's start acting acting acting acting acting