E350 | Dan Bowyer ⁠and  ⁠Mads Jensen⁠, SuperSeed: SuperSaaS Insights, Tech & Global Progress, CBDCs, US Immigration, and The Future of Open Source AI

0:28Well, delicious, delightful. step change and quite interesting uh the the central bank digital currency i know that's something that's on your mind that looks quite interesting the cbdc um the the nvidia acquisition of run ai has caused a bit of consternation in the states uh understand what's happening with with uh with purchases or acquisitions at nvidia founder mode who's not talking about founder mode immigration and the future of AI. So open AI, is it under threat? So full, full, full docket. Mads, where should we start? Start from the top. Let's do it. So quick review of SuperSaaS. So for those that haven't seen SuperSaaS, we've been fairly noisy about it, but I'm sure many haven't seen it.

1:14It's an event that we run for founders to connect them with corporates to understand how to raise venture. So building, selling, raising is the condensed accelerator in one day. And I wanted to just talk about ultimately getting to how do we improve it, what the founders want. But some key takeaways for me were Matt and Nobodon from System talking about their very sensible, smart way of working at how you position your startup. So looking at your homepage asking yourself now you can the first words that fall out of your mouth should be the lead line on your website so that was one of the key takeaways for me um we did a panel with some um some corporates some guys from procurement um from mars and from bmw and there was other big corporates there at the event um and one of the key learnings i took away was how much they value relationships and to me dealing with big enterprise is always the machine but actually these guys and girls really value relationships and building relationships.

2:17And there's something from Andrew from our team. He was listening to Stuart Bewley. Now, Stuart is a great communicator and is an ex-actor and teaches people how to present their stories really, really well. And he shared the fact that it's only 7 % of what you say, the content of what you say creates the impact. 38 % is the tonality and 55 % is your body language, how you present. So they were my key takeaways. Mads, what were your key takeaways or how did you find the day? I thought it was a great day. I think for me, the key takeaway is that we have some incredible, incredibly sophisticated corporates all over the world and also in Europe.

2:59And sometimes they can come across as bureaucratic and slow moving, but they're very powerful. They have amazing distribution. They have so much muscle behind them. And they are looking for innovation. They want the next big thing. and they're just looking for smart ways to buy it and integrate it into their operations in a way that scales. And so they value these opportunities to meet the young innovators, to meet the young companies and figure out what is it that's moving, what's shaking, and what can we really pick up and adopt and use and leverage in a meaningful way. And so for me, just the fact that it is there, it's real, and the corporate's valued so much, that's always reaffirming for me to see.

3:35Yeah, I think there's something very special in the events and founders watching, if you haven't joined in and you're doing something in b2b that couldn't be sold into the into these big corporates please have a look at it and see if it's something that that would that would fit we're running our next one in december um and it's all like i said it's all about this how to build how to raise how to sell and we bring in some some brilliant people and connect you with potential customers and they tell you how to have a sell to them um next on the docket is the world is not turning to shit, Alex Stapp posted on X.

4:08And I love this. I don't know if you wanted to kick this one off, Matt. Yeah, no, look, we talked a little bit about it last week, how there is, it's easy to get this sense when you open the news, whether it's your favorite mainstream or whether it's X, that everything is terrible. And yes, there are challenges in the world, no doubt. But if you look on a grand scale over the last hundred years, there's just been such and incredible progress in living standards, whether we look at poverty, education, literacy, democracy, vaccination rates, or child mortality. So sort of all these big things we look at society, this kind of lenses we look at society through to work out, are we actually making progress?

4:47And the truth is we've made incredible progress. So if you go back to the 1820s, you had 79 out of 100 people on the planet were living in extreme poverty. Now, right, that's extreme poverty. So nearly 80 percent. Today, that's down to less than 10 percent. Is there still too many? Yes, but we made incredible progress, especially since the 50s. Similar for education, if you go back to the 1820s, 17 out of 100 had some basic education. Today, that's up to 86 percent. Literacy, 12 percent were able to read 150 to 200 years ago. Today, that's up to nearly 90 percent. So there's progress on so many dimensions and it's just important to remember how much progress we're making as a species of this planet yeah the news obviously likes the if it leads if it bleeds it leads kind of mantra and um with all the political turmoil and every time you open x or every time you open up the bbc news app it is doom and gloom and disaster but it's also out of context i think it's a really interesting interesting reminder so that there was the our world in data stats that kind of just reminded us of that.

5:58It's been around on the social web for a while. The one thing I wanted to just kind of raise, maybe there's some reflections on this, is technology has had a massive play in all of these poverty, education, literacy, child mortality, vaccination. The only place that technology hasn't really hit and the only place where it hasn't really had that much change or that much impact is in democracy. So roughly half the world is living in authoritarian regimes and half the world is in a democracy. And it's the one area, I would argue, where technology hasn't touched. And I wonder if that's correlation causation, or if there's any reality in that.

6:34What's your reflection? Well, it's a good question. I mean, if you look at the stats going back 70, 80 years to the time around the 40s, the Second World War, very, very few countries were living, very, very few people were living in democracies. And today, it's about 50 % of the planet. So there has been some progress, but there are also cultural differences. Maybe democracy isn't for everyone. I think there was this Western experiment of trying to impose democracy in some countries in the Middle East 20 years ago, 15 years ago. That was disastrous. The invasion of Iraq did not turn out very well at all.

7:09I think we've had attempts at implementing and imposing something that looks like democracy in various African countries for a long time. In many places, is it's just not going very well. Is that because democracy is a flawed concept? Is that because it's not for everybody? Is that because of other underlying factors? I mean, there are many theories around it. I do think there has been progress, but it's certainly not evident, self-evident that democracy will be something that is implemented everywhere and for everybody. Maybe the same way we're thinking about some of these other things like literacy and education, et cetera.

7:45Yes, democracy is problematic. I guess is the short answer. Moving on. So the US, EU and UK have signed the Council of Europe's Convention on AI. If you haven't come across this new AI regulation, I'll just give you some of the top lines. So the US, EU and UK have signed as a legally binding international treaty focused on AI regulation. It's mostly focused from what I can see on human rights and democratic values. the treaty requires signatories to ensure there is legal recourse for victims of ai related ai can't speak ai related rights violations it took two years and was agreed by more than 50 countries critics have poked it a little bit and said it doesn't have enough teeth um there are a lack of sanctions based in the regulation as it stands it very much aligns with europe's ai act the G7 deal, the Bletchley declaration.

8:41So there's an alignment with other regulatory pieces. And Vera Jarova, the European Commission's vice president, stated that the framework strengthens trust in AI and aligns global efforts. So 50 countries took two years to put together. And it's the starting point. What do you think? What's missing? What's working? Will it work? What do you reckon? Look, if you believe, as I do, as we do, that AI is important, it probably follows that we need some kind of regulation. Rene Godreld is right. There are so many considerations. I think the question is that you picked up on this question around teeth.

9:21Does it have teeth? It does have a sense that it sort of obligates the treaty signatories, which is kind of the main ones right now, US, UK and the EU. It obligates them to put national regulatory frameworks in place that puts some burden on themselves and on companies operating in those countries. But let's pick up on a few things like data privacy. Data privacy is one of the important things that is pulled out in the treaty. But I can't see any chance that data privacy will be respected. and we all already have massive data drag nets and ai makes it much easier to troll those for governments and i think governments will do this no matter what they say i think privacy is essentially dead as a concept now we can discuss whether that's desirable or not but i don't think we can discuss the direction of travel and so to some extent i don't know whether a treaty like this is sort of just almost pretending that you know you're pretending you're doing one you're saying you're doing one thing and pretending, but the reality is probably going to be different.

10:27The second vector, second thought I have is around the competitive posture, because it is unlikely that Russia or China will sign this. And even if they did, they'd ignore it for sure. So as with everything we do, we should ask ourselves, if we put rules upon ourselves that slows down the development of AI, and if our competition are not bound by the same rules, and if this means that they will end up with more powerful AI than we have, what will that do for our freedom and our human rights in the long term? It's all very well to say that our national government shouldn't spy on us. But if that means that we just have foreign adversaries spying on us and taking our data, stealing our trade secrets, are we really that much better off?

11:15I think there's no simple answer yet to how we should regulate AI. I think it's probably helpful that this work has started. I question kind of the medium term efficacy of this. I think this is something that will have to play out over some years as we get smarter. And we certainly must consider the trade-offs at every step. My sense is that this is a kind of helpful guardrail prelim to gen AI. So I get you. There is no way to, I mean, the genie probably has been out of the bottle for some time and AI is just an accelerant. I don't believe privacy has been around for a while. I mean, you could argue since the advent of the internet or social media, but it feels like we need a starting point.

12:08so i'm i'm warmed that this that the us the uk and the eu are are conjoined in this effort i i welcome the guardrail starting point and i also get this kind of prelim to potential you know we don't know how far away gen ai is but it might be it might be it's sensible to have these conversations but let's see let's see how it unravels leading on from that i i think it's important i know i think it's important i think having some level of national responsibility for oversight and awareness is important i mean effectively we have just incredibly powerful models being developed and there has to be some checks and balances it's it's unfathomable that this will just be self-regulation i just i can't i can't see how anybody who believes that ai is powerful believes that you can just leave people to regulate themselves we wouldn't do that in any other sphere.

13:02I mean, imagine, you know, chemicals or foods or, you know, energy. We have regulation when things are powerful and this is so powerful. So there has to be something, but it's certainly an open question mark as to exactly where we draw the line. I just think that people over egg the power of LLMs. I think it just feels like we've got this really powerful autocomplete these chat gpt these lms and i get it there's there's so much to be done and there's like high expectations but there's an off switch you can pull the plug this is not i don't believe there's as much power yet how far away is that power to be discussed but i just can't i don't buy that there is that much death and disease and destruction to be created by these by these toolkits as they stand as with everything we overestimate in the short term and might underestimate in the long term i mean it's only a week ago we were talking about autonomous bots here in this podcast um one of the things we've seen is that the microservices architecture has proliferated in medium and large size enterprises over the last decade these are incredibly complex architectures if you look at companies like netflix or spotify they might have tens of thousands of microservices that they use to run their infrastructures.

14:23Now, imagine that we turn all of those into autonomous agents. It's kind of the clue is in the name. You've got agents that can do something and act in a way that's not prescribed. Now, where will that go? We don't know. And it's logical that the vast majority of things we see are going to be completely harmless, but we are just developing something that's very powerful. And we're seeing a trajectory towards allowing more of this technology to operate independently of constant human supervision and guidance. And so where could that lead? A medium term, probably nowhere very dangerous. Longer term, who knows?

15:05We know we are still in the infancy of how powerful these models will become. If we look at the amount of training data that the most sophisticated LLMs have been trained on today, it's in the order of maybe a 10 ,000th of a 20 ,000th of what it will be once we unleash the total training data set we have. All the YouTube data has barely been scratched in terms of using that to train LLMs. So there's so much further we can go. These models can become so much more powerful. So I'd say, yeah, probably overestimate in the short term, medium term to long term. I think this is going to become a really big thing.

15:48Yeah, and I'm always intrigued by the amount of energy required to create or to power these models. We're going to have to have an energy revolution as well to power all these chips. Well, the good news is we have had an energy revolution, right? If you look at the cost of solar, the learning curve, the efficiency gain is massive. Batteries, we're seeing constant improvements. There's so much happening. Now, we happen to live in kind of the cold north where solar might never become kind of the predominant energy source. And we have other things. We have wind. And we're probably going to have to put some nuclear into the mix.

16:25But in our lifetime, we're going to see essentially net zero energy paradigms. Possibly the only thing we won't be able to use electricity for is maybe space launches. But everything else with the battery technology and with the advances in PV and other renewables, I think we are going to see a dramatic improvement in energy production. So, yes, you're right. Again, shorter term, we're a little bit stiff. Again, the Chinese, they've stolen a march on us. 300 nuclear power plants under construction. They're going to have a massive competitive advantage in AI because they're going to have the energy to fuel the data centers if they can get hold of the chips.

17:07Whereas we have been way too slow with permitting and planning some of the nuclear facilities we need in the short to medium term to power this. But longer term, there's a very, very optimistic picture. It's super interesting how it's going to unfold. And it's going to be so quick. Now, next up, we've got the digital euro. We've got the central bank digital currency. So the idea here is that ultimately a digital euro will make payment services less costly for the public. That feels like the outcome, which can only be a good thing. This is obviously not crypto. This is digital fiat, not the crypto nonsense.

17:48Where were you going with this? What was on your mind? I think it's incredibly important. It's maybe been a little bit underreported. I think some people aren't quite understanding the magnitude of what this can be and become. Let's run through maybe some of the background. So the whole idea is that the European Central Bank is planning to launch a digital euro, which is effectively a digital currency. So it's like cash, but in a digital wallet, a little bit like cryptocurrency. And the idea is for it to complement cash rather than replace the physical cash. So physical cash will still continue to exist.

18:23But as we know, most of the money we use today is not physical cash. It's money that sits inside commercial banks. It's claims we have on deposits. And when we spend money, it's effectively a claim we have against one bank where we might have a deposit that then is shifted as claim to another bank. So it never really sees the real world. The digital euro would completely upend that paradigm because it's no longer a claim against your bank. It's a claim against the ECB. So that makes it safer because there's no default risk from the commercial bank. You're effectively just transacting money that you have a direct liability relationship with the European Central Bank for.

19:06The idea is that it will be available to everybody through these wallets that are provided by the ECB. And that means you can get access to a bank account without needing a kind of this digital currency without needing a traditional bank account. So there is an accessibility issue. And then you have the whole notion of instant payments that are not running on the traditional payment rails. So this could have huge implications for the cost of making payments and for payment service providers. Now, some considerations here. There's some privacy concerns that have been concerns that it would allow the European Central Bank to snoop on all transactions.

19:42It's still being explored how to balance privacy with the need to prevent illegal activity like money laundering. And so the notion is it will likely offer more privacy than current electronic payments, but less than cash. And so because of that, there's been some concerns. Of course, there are lots of folks still that like to transact directly in cash that don't like the notion of people being able to look at what's being spent. There are also questions around the impact on banks. Today, our banking system is running on this notion of kind of fractional reserve banking. What does that mean? It means that the banks that provide all the loans to businesses and mortgages, et cetera, they're effectively lending out the money we have deposited.

20:25So we get our salaries, we deposit them in the bank, and then the bank turns around and lends them out to somebody else. and they take a margin cut there, right? They'll pay you 2%, they'll charge somebody else 5%, and that difference, that's their profit. Now, what happens when currency is issued directly by the ECB, and you don't have the, it's not just a claim on the bank anymore? That could really change how banking works. So there's some concern in banks about the future of the banking system. And then there are, of course, lots of questions around the technology itself. Will this run on blockchain?

20:59will this be a centralized database? And some of these things are still being worked through. But I think, listen, it's still relatively early days, but I think it's incredibly interesting and this could really truly change the way money functions in Europe. So watch this space. Do you know where they're at? Do you know what the, any release, any status? Yeah, so we have here in the fourth quarter, we have the next major update, sort of a go-no-go picture where effectively there's been lots of studies and surveys being done and analysis and they're going to make kind of the next big decision milestone decision about whether to move forward with this initiative or not or not so this is this is upcoming in just a few months away i know that there was a part of the report it was saying that um when they look back through historic data they could they ascertain that european payment companies stock prices would go up as american payment company stock prices would go down so i don't know how they did that but it's quite an interesting it's quite an interesting outcome um next up let's get into founder mode paul graham has has uh written a paper um chatting with brian chesky and chesky looking at steve jobs and founder mode versus manager mode and the old paradigm of investors like us saying to founders as they raise capital and as they progress through their journey to get really strong functional people into roles and then set them free to do their thing and obviously Chesky was saying like 16 didn't like the business 17 18 19 20 was really struggling with placing and getting into manager mode and then took back control of the reins as a founder mode is effectively micromanagement was what came to mind but that's not the right turn of phrase and then it made me think about jensen jensen from from nvidia has got allegedly has 60 direct reports and has been living in founder mode for possibly decades so what's the deal here do you think and i want to talk about steve jobs's retreats how he how he came back to apple and then and then effectively got into founder mode and i was also looking at nival's commentary on And he was basically saying real founder mode is just not giving a damn, which I thought was just a really nice, simplistic overview.

23:19But where do you want to go with this, Matt? It's founder mode. Over to you. Look, we all see it every day. The founders that are the best are those that have an awesome command of their field, that have a vision, that are strong leaders and can drive themselves and their organizations forward. And if we look at the most successful CEOs and founders today, if you look at Elon and Jensen Wang, what makes them the best CEOs in the world? And I've seen lots of different analyses, but three interesting things to pull out. One thing is they always work on the most important thing in the company. They have maximum focus for maximum leverage.

23:57If there are production issues at Tesla, Elon will go and live and sleep on the manufacturing floor until that problem has been solved. Jensen Wang is famous for kind of, you know, just going and setting his desk up in whatever department that has an issue right now. And that's where he works. And that's where he focuses. No standing meetings, no reviews, no committees, no proper reports on a kind of biweekly basis. No amazing meeting. there. But exactly, he is there at the point of maximum impact. And I think there's something really powerful about that. And I'm not sure it's a new phenomenon.

24:35I think we've known it for many years. I think it's called leading from the front. It's the leader being where the action is and helping the team solve the most difficult problems by providing that leverage. And so the second thing is they love hearing bad news. They want it directly from anybody in the organization. if something is broken, if something is not working, they want whoever has spotted that to come and tell them so they can make sure it gets fixed as quickly as possible. That's very powerful. Many people, you know, as many people, we have a natural adverse reaction to bad news. We don't like it.

25:08We want to pretend that it doesn't happen. And what I think can happen, especially in large organizations, is you put a lot of layers between you and the bad news. You can almost pretend it doesn't exist. But what they do so powerfully is they gravitate towards it. And back to the first point of point of maximum leverage and focus, you find the bad news and you focus on it, you can solve it more quickly. And the third point is they are completely mission driven. We spoke a lot about that at SuperSaaS. If you look at Elon, his companies are so mission driven, right? PayPal, make local commerce frictionless.

25:42Tesla, accelerate the world's transition to sustainable energy. SpaceX, make mankind multi-planetary. I mean, these are huge, audacious missions. And that attracts a very special kind of person, a very special kind of employee who's got that mission-driven seal and who's willing to work, do whatever it takes to make things happen. Those three things are just incredibly powerful. And they are not for everybody, neither as employee nor as, frankly, as founder. But I think what we've seen is the founders that build the best and strongest and most enduring companies, they all have some element of these three aspects.

26:21What do you think, I mean, those three aside, what do you think is the key learning? Obviously, lots of founders are reading on this all over the social web. What do you think is the takeaway for an early stage founder reading about founder mode? You've got to really want it. Jensen really, really wants it. He is an absolute cockroach and he has sacrificed everything. He has taken the company needed to the brink of collapse and bankruptcy multiple times. And he's always found a way. He talks about pain and suffering and whatever it takes. And that's not for everybody. Yeah. In fact, I'd say many of the founders we need, it's pretty obvious quite quickly that they You don't necessarily have that absolute seal.

27:10That doesn't mean you can't start a good company. That doesn't mean you can't build a good business. It's just a different company, right? But you probably never build one of the global titans if you don't have some of that seal. Sorry. No, I think my key takeaway, it's a nice idea. Obviously, Paul Graham isn't creating anything new. He's just creating a window or a lens onto certain people's behavior. And it just it did remind me along the lines of what you were saying, just reminded me to remind founders to be candid about what kind of business they want to build. So if you want to build a venture scale, global speed of sales, speed of scale business, this is probably the attitude you want to consider.

27:56Doesn't mean you don't hire really super smart people and enable them, but it just means you create these direct report lines. And if you don't want to do that and you don't want to build a venture discovery, that's okay. It's okay to put a really substantial 10, 20, 30, 40 million pound business. That's an amazing, incredible thing. I think you're absolutely right. I mean, two things to pick up on there. I think if you think about founders that are incredibly focused on being in kind of the point of most impact and really being involved, well, that means they won't be able to do some of the stuff that other managers are.

28:33kind of all the reviews and looking at everything. And that puts so much more onus on the other leaders in the business to be really good at making things run because they're not going to be babysat in the same way. So if you're not in that part of the business that is the most strategic thing right now, you're probably going to be left alone a little bit to figure it out. And that means you need very, very high quality of people around you to do that. That's point one. Point two, I would say there's the mission-driven thing, which is that nobody likes to be micromanaged, but sometimes we can forgive it if we think the mission and the leadership and the progress is so great.

29:11And there's one thing I've always seen is almost everybody likes to win. And so if we're winning and if we're feeling that the leadership is taking us in the right direction, even if we might be annoyed by some of the habits and some of the ways it's done, we can forgive a lot if we're winning. So some considerations there, I think, just to throw into the mix. Yeah, no, totally fair. Last up, I wanted to cover with you, we've talked about freedom of movement in Europe and looking at the reports on the US. Since 2020, the US has experienced its largest immigration wave in generations was the title in the Wall Street Journal.

29:54now we we've talked a lot about i mean the the uk has just denied the eu's request to re-implement or re-enable young persons freedom of movement so the 19 to 25 freedom which just makes total sense and we've talked about this before we hope they'll reverse that decision but but immigration as you've talked about before is a is a great powerful force so where were you going with this So I think it's a massive topic, but my lens was how is this going to affect Europe? So looking at the American situation, what lessons are there for Europe? What lessons are there for the UK? Well, I think there's sort of an element of connecting the dots.

30:38So the US has been growing faster than Europe over the past couple of years. One of the reasons is that they have been investing a lot. They're running very large fiscal deficits. But another reason is immigration. And one of the things that's not always acknowledged is that, yes, we have had productivity issues in Europe. Productivity has slowed down over the last decade and a half. The same has happened in the U.S. If you look at the period since 2010, U.S. productivity growth has been dramatically underperforming compared to the 20 years period prior. And the main reason the economy has continued to grow as rapidly as it has is because of immigration.