The race isn’t over for European AI startups, according to Accel Parther Philippe Botteri

“It's a bit too early to say that the race is over,” said Philippe Botteri when asked about European startups’ AI progress. “I think we're just at the very early innings of this race.” Botteri is a partner at early-stage investment firm Accel with over 13 years under his belt at the firm, leading investments in DocuSign, UiPath and more recently Snyk and Chainalysis. Today on TechCrunch’s Equity podcast, host Rebecca Bellan caught up Botteri to dive deep into Accel’s Euroscape 2024 Report. Tapping into Botteri’s experience in Cloud, SaaS security, and enterprise sectors, the pair discuss AI's rising influence, its impact on software and cloud investments, and how European startups can compete with the US. Listen to the full episode for more about: How AI is eating the software market, with AI and cloud funding predicted to hit $79.2 billion by the end of 2024. The challenges faced by traditional software companies as funding growth slows outside of AI. Why Europe’s strong talent pool gives it an edge in the AI race, even as startups on the continent struggle to compete with the ungodly amounts of money U. S. tech giants have. Increased M&A activity globally amid a slow IPO market. Why 2025 will be the year of the “agentic revolution” with AI significantly impacting software development and productivity. Equity is TechCrunch’s flagship podcast, produced by Theresa Loconsolo, and posts every Wednesday and Friday. Subscribe to us on Apple Podcasts, Overcast, Spotify and all the casts. You also can follow Equity on X and Threads, at @EquityPod. For the full episode transcript, for those who prefer reading over listening, check out our full archive of episodes over at Simplecast. Credits: Equity is produced by Theresa Loconsolo with editing by Kell. Bryce Durbin is our Illustrator. We'd also like to thank the audience development team and Henry Pickavet, who manages TechCrunch audio products. Learn more about your ad choices. Visit megaphone.fm/adchoices

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:11Rebecca Bellan?

  Hello and welcome back to Equity, TechCrunch's flagship podcast about the business of startups. I'm Rebecca Bellan and this is our Wednesday episode where we home in on a trend in the tech world and dive deep. Today, we're talking to Philippe Botary, the Accel Partner who's led the firm's investments in companies like DocuSign, UiPath, and more recently, Sync and Chain Analysis. Philippe, welcome to the show.

• 0:33Speaker 01

  Thank you, Rebecca. My pleasure to be here.

• 0:35Rebecca Bellan?

  I'm really excited to have you on the show to talk about all things Europe and AI and cloud and SaaS. You're based in London, is that correct? But you used to work at Bessemer?

• 0:45Speaker 01

  Yeah, so I've been in Europe for the past 13 years now at Accel. Before moving back to London to join the firm, I spent a bit less than 10 years in the US and a good part of that time was with Bessemer. I was in their middle part of us.

• 1:00Rebecca Bellan?

  And so now you're focused on cloud, SaaS, security, enterprise IT. Has the conversations around these technologies shifted very, very much in the last decade or however many years? Or has it kind of been similar conversations with different jargon?

• 1:16Speaker 01

  So yeah, I'm spending a lot of my time these days on cybersecurity and AI. So it has been incredible in the past 13 years to see how things have changed in Europe. I mean, when I moved to Europe in 2011, journalists were asking like, when is Europe going to generate the first unicorn? Like at the time, the unicorn was not pinned, but it was like, when is Europe going to generate the first billion dollar outcome? That was the big question. In 2021, UiPath was the largest software IPO globally. And who would have said that the largest software IP of 2021 would come from the company out of Bucharest, Romania? It tells you how far the European ecosystem has gone in the past 10 years. And from an investment standpoint, if you look

• 2:01Speaker 01

  at the venture capital going into software and cloud companies, you know, in 2016, it was one tenth of the equivalent in the US. And today, we're one third. So it shows you proportionally how far Europe has grown as well.

• 2:18Rebecca Bellan?

  And now your firm just recently came out with, it's called Euroscape Report, with a lovely image of AI eating software. I would love to just hear some of your high-level thoughts about some of the findings of the report.

• 2:31Speaker 01

  Sure. The reason why we call this report AI is eating software is because AI is really rewriting software, both figuratively and non-figuratively, right? We have models who can actually write software code. And if you look at what's happening in the software world, AI is basically driving the recovery of the ecosystem after kind of the decrease and the 2021 bubble. And that's true in the the public markets. And if you look at the public market, just looking at the NASDAQ in the past 12 months, there was like little more than 8 billion of value being created. And 5 billion of that 8 billion is coming from a set of six companies.

• 3:14Speaker 01

  And I can name them, but you know where they are, right? You know, the Facebook, Google, Meta, et cetera, and Nvidia. So I don't say the magnificent seven, because, you know, I think Tesla is missing in action. Their performance the past year hasn't been as great. But the six companies have taken a lot of the values being created in the past year. And if you look now at the private market and the money going into private cloud companies, this year it's around, we estimate it's around $80 billion, which is 65% growth versus 2020. The entire growth is coming from investments in AI. I mean, the amount that we're talking about for AI, I mean, the past couple of years,

• 3:58Speaker 01

  you're talking about 56 billion. So you can call it around 25 to 30 billion a year for the past couple of years. So huge investments. And I think it really highlights the potential of the technology in terms of productivity improvement and its ability to basically redefine the capabilities of software.

• 4:19Rebecca Bellan?

  Yeah, I think actually TechCrunch had an article a couple of days ago from Kyle that investments in Gen AI startups topped, it's close to four billion in the third quarter alone of 2024. But you know, the report mentions that there's growth slowing outside of AI. Do you see that as an opportunity or as a risk?

• 4:36Speaker 01

  Well, I mean, I think what's happening in the AI world, everything is a bit of a frenzy right now, is a lot of investments and growth. Now, if you look at the traditional software companies, then the picture is not as bright. And I think what we're seeing right now is the growth rates, we're following a public index of a hundred or so public cloud companies. That index on average was growing at forty seven percent at the beginning of 2021. Now it's moved from forty seven percent to fifteen percent in the past quarter. So it shows you that AI is really sucking the air out of the room. And it's quite hard for the traditional software companies to basically be able to grow.

• 5:19Speaker 01

  So obviously, they are focused now more on profitability. And they're working hard to incorporate AI models into their software so that they can start to actually grow faster. So we'll have to see what what happens in the next couple of years. But I think if I try to explain this, it's a combination of different factors. I mean, the one is there's been a lot of spans of big companies buying software in 2020, 2022. And now they're kind of digesting these spans. Two, there is kind of a rationalization, like a lot of things were both. And companies are saying, well, maybe I don't need to have five different communication software. And I can kind of unify on one. So there is also consolidation of spend on the big platform.

• 6:02Speaker 01

  There is AI, who is basically where we're seeing IT budgets are now being freed up to be invested in AI. And so obviously, even if they are growing, they're growing three, four percent. But the investment in AI are growing much faster. It's like thirty percent. And so that means that the rest is actually decreasing. So that's putting more pressure on traditional software. And obviously, you have the macro uncertainty that giving opposed to companies or sales process are also slower.

• 6:30Speaker 01

  I think that's kind of what we see happening in the traditional software world.

• 6:34Rebecca Bellan?

  Yeah, and then well, Europe obviously is very well known for its enterprise software, specifically more so than consumer. Does this mean Europe specifically is in trouble if all of the US AI foundational models end up eating enterprise apps?

• 6:46Speaker 01

  Well, right now, the US has the advantage on AI. I mean, if you look at the amount of investment going there, you know, if you have $10 invested in AI, eight is going into the US, two is going into Europe. So, yes, there is more money coming into AI in the US. But Europe has a lot going for it in terms of AI. Like, if you look at the talent pool that we have in Europe, it's a pretty strong talent pool. I mean, Yann Le Cun opened Meta Fair in Paris in 2015, right? They opened it there because they know that it was a pretty strong pool of talent there. And France has a very strong history in terms of metal fields and applied math, and that translates very well into machine learning and AI.

• 7:30Speaker 01

  Then you have DeepMind in the UK with a very strong talent pool. You look at stable diffusion, which is probably one of the most used algorithms to create images, and that was invented in a lab in Munich. So I think Europe has a very strong talent pool. I think it's fair to say that on the foundational model layer, they started a bit later. So you have some great companies like Mistral, like H, like Poolside, Eleven Labs, but they're a bit behind. So in terms of their fundraising capabilities, they have raised hundreds of millions, they haven't raised like the billions that OpenAI, OnTropic have raised. But that is going to be the next step for Europe, and the question is, you know.

• 8:11Rebecca Bellan?

  What you think the next step is to get to, I don't know, invest into European foundational models to not rely on US models?

• 8:18Speaker 01

  Well, I think there's a question of whether these companies will be able to raise billions of dollars.

• 8:23Rebecca Bellan?

  Right.

• 8:23Speaker 01

  Even if they don't, I think there is room for these companies to basically focus on more specialized models that can provide value. I mean, if you look at enterprise application in particular, two things are very important. One is the inference cost, two Is a latency. So thinking that the biggest model with the highest latency and the higher inference costs are going to solve everything. You know, I don't think it's going to happen. So I think there is room for this AI would call challenger to actually create a lot of value. Now, I also think that some of them will actually end up raising a lot of capital and potentially change leagues.

• 9:03Speaker 01

  And it comes a leagues of companies that are raising billions, which right now you have Entropic, X and OpenAI.

• 9:11Rebecca Bellan?

  Right.

• 9:11Speaker 01

  And then you have the next league, which are, you know, the, the, the big companies, the Google.

• 9:16Rebecca Bellan?

  Yeah, the public companies that spending hundreds of billions of dollars on their AI.

• 9:21Speaker 01

  Billions. A lot of it goes in infrastructure as well. And so the question is, is infrastructure going to be, are we getting to a point in the future where the best models will require tens of billions of investment, in which case then it's going to be challenging for the others to catch up. Because, you know, when you're worth two or three trillion, you can spend tens of billions. I think it's hard for anyone else.

• 9:45Rebecca Bellan?

  Hmm.

• 9:46Speaker 01

  But I think we'll have to see because I think as model get more efficient, you can train more powerful model without the same increase in compute capacity. So we'll have to see how that's going to play out.

• 9:57Rebecca Bellan?

  Yeah, owning the infrastructure seems like it'll be very important. And, you know, it seems like a lot of these companies are even just becoming asset companies. At the end of the day, they have to own the infrastructure, they own the data centers. And at this point, it looks like they're owning a lot of the energy production as well as we see companies like Microsoft invest in nuclear power and things like that.

• 10:16Speaker 01

  Yeah, the Three Mile Island deal where, I think they purchase the production for the next 20 years or so. In the past couple of weeks, both Google and Amazon announced deal for SMR, so small nuclear reactor.

• 10:30Rebecca Bellan?

  Yeah.

• 10:30Speaker 01

  To be able to power their data center. So clearly there are strategic moves being made to secure that energy production.

• 10:38Rebecca Bellan?

  Yeah, so how do startups, you know, as you call them the challengers, how do they even bother competing? I mean, I don't think XL invested in this company, but correct me if I'm wrong. Aleph Alpha, which last month seems to have pivoted, they're a German company. They had raised over 500 million and they were trying to do their own LLM and build their own foundational model. But I think just despite the fact that they raised 500 million, that just simply isn't enough funding necessarily. And it's difficult to attract investors, I think. So they pivoted to doing kind of AI support. And then they do still hope to develop their own advanced generative AI model separate from Silicon Valley. But yeah, I think, as you said, a lot of these companies are going to have to focus on smaller use cases if they were late to the foundational model race.

• 11:21Speaker 01

  So for Aleph Alpha, we didn't invest in them. I think I'm not close enough to understand exactly why they did this pivot. But I mean, if I look at Mistral and H, they have plenty of capital. They are building their foundation model and they're not slowing down. So I think it's a bit too early to say that the race is over. I think we're just at the very early innings of this race. And I think it's just too early to call it at this point.

• 11:47Rebecca Bellan?

  Yeah, it's just if you do need those levels of funding, where does Europe turn to? Where do European challengers turn to?

• 11:53Speaker 01

  I think it is a fair question, but if you look at where the money is coming from and you have obviously private funds and both if you look at the investor in Mistral and H. I mean, it's a very international base. So I think on the venture side, I don't think that the US has an age. I mean, their same funds are investing in Europe. And so there's access to the same capital pool. I think if you look at the big strategics, the Amazon and Google and Microsoft. I mean, they have made investments in different companies and they're obviously in contact with the companies in Europe and would be happy to do it. So there is an option where the capital is coming from the traditional venture funding from this strategic.

• 12:34Speaker 01

  I think there is also an option where the funds are coming from the sovereign funds. I mean, we have seen a lot of interest, you know, in the Middle East for AI. So I mean, the money could come from there as well. If you look at the evolution. Of technology, I think what it has shown is that if there is an interesting opportunity, you'll find the capital. I think the key is to make sure that there are enough progress being made. So people really see that there is a big opportunity for this company coming out of Europe. And then if there is, I don't think access to capital will be an issue. I think it will be an issue if people think that they cannot succeed.

• 13:12Rebecca Bellan?

  Well, how do you change that mindset, I guess, or what strategies as a VC do you use to kind of promote your portfolio companies on the international stage and ensure they're getting access to those pools of capital?

• 13:26Speaker 01

  Well, I think at the end of the day, it's the team and the product that speaks for themselves. So I mean, if you look at the different model that Mistral have released, I mean, it didn't take very long for people to take a very strong interest in them because they were just very good and very performance. And so H has not released their models yet. I think probably going to come between now and the end of the year. And I think the companies will have to look at what they produce and whether they they think it's exciting and whether they think it can basically start the agentic workflow revolution that people are waiting for.

• 14:00Rebecca Bellan?

  Ah, yes, the agentic workflow revolution.

• 14:03Speaker 01

  Ontropic released something today

• 14:06Rebecca Bellan?

  Yes.

• 14:06Speaker 01

  About the new agents that they launched. But I mean, H is not the only one working on it. I think Microsoft also.

• 14:13Rebecca Bellan?

  Microsoft also, yeah.

• 14:14Speaker 01

  You know, also announced.

• 14:15Rebecca Bellan?

  Salesforce.

• 14:16Speaker 01

  So everybody, I think, is going into this direction. But right now, I think it's been more talking about it and announcing release that really providing tangible results. So I think it's going to take a few more months for that.

• 14:30Rebecca Bellan?

  One thing I did want to mention, because you were talking earlier about the talent pools and how Europe has great talent. And that is one of Europe's strengths, I think, when it comes to, you know, not falling too far behind in this AI race. But then it's worth calling out that big tech has consolidated AI talent, you know, using their their billions to tens of billions of dollars. There's the character AI founder who went back to Google, I think. There's the inflection Microsoft deal. There's the adept. Amazon, there's these reverse acquihires, if you want to call it that. How vulnerable are European startups to falling under the umbrella of these larger big tech giants?

• 15:09Speaker 01

  Well, I mean, I think somehow if you go back to where you came from is probably because something was not developing the way you were expecting to, right? So as long as companies are successful, there is no reason for them to go back in the fold to some extent, right? So all the people like the great scientists who have started these companies in Europe, they all came from DeepMind and Meta, et cetera. So why would they go back there? So either, you know, obviously if there's a lot of money at stake, et cetera, like who knows? Potentially, they may do that. But I think if you look at them today, I think my sense is that they're really on a mission.

• 15:53Speaker 01

  And the reason why they left these big companies is because they saw that they could achieve what they wanted to achieve better in a different environment. And I think a lot of what we heard is like when you're part of big research organization, et cetera, you have less freedom to develop things, to ship things. And they basically decided to leave and create their own company so that they could go back to smaller environment where they can be a lot more productive and efficient and with less constraint around them. So right now, they are on a mission and they want to succeed. And so if at some point, something doesn't pan out the way they wanted, yeah, maybe they go back. But again, I think that's not the pattern that we're seeing now. So we'll have to see how things, and probably some will.

• 16:38Speaker 01

  Like I think, you know, it's a law of numbers. Some hopefully will be successful and standalone independent companies and some will be acquired of some form of hack we hire because of all the regulatory constraints.

• 16:51Rebecca Bellan?

  Okay, and we can probably just keep talking about this, but we do need to cut to break. And when we come back, let's talk about the state of the exit market for European startups and generally.

• 17:05Rebecca Bellan?

  I think that your report also did mention it had some data about M&A activity, the state of the exit market in Europe specifically. And I think M&A activity was expected to reach somewhere around 75.4 billion for 2024 because the IPO market is continuing to stall. Can you talk a little bit more about any M&As that you're expecting to happen this year?

• 17:27Speaker 01

  So the M&A numbers, the 75 billion is a global number. So I think it's what

• 17:32Rebecca Bellan?

  Oh, global. That's not just European?

• 17:33Speaker 01

  No, it's not European. I think it's what we see on a global basis. And that's for M&A deals that are north of half a billion. So it's kind of telling you what's happening in the big exit landscape. If you look at what's happening this year, I think it's kind of in line with the past few years, and it's very driven by very big exit. We say this year is going to be 70, 75 billion, and 35 billion of that is the acquisition of Onsys by Synopsys. Just one deal is half of the market, and that the pattern that we have seen in the previous year was blank in 2023, which was a pretty big one as well. Now, if we look at what's happening in Europe, I think a lot of the names that we are seeing

• 18:13Speaker 01

  here are US names. So we have been probably a bit less of this large M&A happening in Europe recently. That said, I think there is a pretty large pool of late-stage software companies that are in the 200, 300, 400 million in revenue range coming out of Europe that just missed the IPO wave of 2021 and are now waiting for kind of the IPO market to reopen. And so my guess is that in the next two to three years, we'll start to see these companies getting to some form of exit, either through an IPO, if the market reopens, or strategic M&A. But the one thing on the M&A side as well is that the big tech are

• 18:58Speaker 01

  missing in action. The obvious reason for that is because of the regulatory scrutiny. It's very hard for them to make any large acquisition without, you know, huge uncertainty. Right? Every time they do something, it's being reviewed. And even if it goes through, that means for 12, 18 months, the company that's being acquired doesn't really know what's going to happen. And it's a very uncomfortable position to be running a company. And when you don't know if you're going to end up at, you know, Google or Microsoft, or if, you know, 18 months from now you'll be a standalone company and kind of have to recover from that. So that makes actually the the risk on both sides is pretty high. And that's why it's not happening. And I think that is obviously impacting the tech M and A market pretty importantly.

• 19:41Rebecca Bellan?

  Well, it is and it isn't happening, right? It's not like happening in the obvious way, but it is happening in those kind of reverse acquihires. Like we'll just, we'll just hire your whole team, right?

• 19:50Speaker 01

  But that that's very specific to, I think we're only seeing that with AI, where a few key talents in research are driving most of the value. I don't think you're going to see this kind of deal with more traditional software company. Because here basically was they wanted to hire back the researcher.

• 20:09Rebecca Bellan?

  You're talking about Character.

• 20:10Speaker 01

  Yeah, I think there were Character and a few of the others that that happened, right?

• 20:15Rebecca Bellan?

  Yeah, Inflection.

• 20:16Speaker 01

  I think there was a Character, Inflection and Adep. So there were the the three of them, you know, Google, Microsoft and Amazon. And it was kind of the same thing where we're not making a full acquisition. We're kind of taking some sort of licensing agreement, hiring back some of the key people. So very structured deal. But we think that is kind of a bit of a one off, you know, which is very driven by the dynamics of fundamental model design.

• 20:43Rebecca Bellan?

  You mentioned also late stage funding in Europe. Now, I think that late stage is just difficult for a lot of companies at the moment generally as investors are perhaps a little more risk averse. And Europe kind of has a reputation for implementing austerity measures since 2008. But are we starting to see? I think we've covered a few articles, like 20 VC closes a new $400 million fund to, quote, "make Europe great again." Or the Nordic entrepreneur-led VC firm Node VC closes 71 million euro in its first fund. We're starting to see more late-stage growth funds come to life. Are you seeing that? Does that pan out to you?

• 21:25Speaker 01

  Well, I mean, I think the fact that we're seeing capital flowing into European funds is because I think the venture community has realized that Europe was a very strong component of the venture ecosystem globally and that companies, great companies, could emerge from the region. I think that's what's attracting the capital. We often hear, oh, you know, Europe or this country or that country is kind of missing late-stage capital. I think in today's world where capital has really no boundaries and funds invest across region, I think that saying that one specific country is missing late-stage capital, I think is totally missing the mark. I think what it says is that the late-stage capital

• 22:06Speaker 01

  doesn't want to invest in some of these companies, which is different, right? Because not every company is investable for everyone. If companies of, you know, very good growth rates, strong ambition, want to be global, they're going to have access to capital. So I don't see that really as a...

• 22:23Rebecca Bellan?

  You don't think like the geography of it all can affect certain people? I mean, I'm just thinking about just you saying that made me think about New Zealand, where I used to live. And I did a bunch of coverage of the venture environment there. And Peter Beck, who's the CEO of Rocket Lab, said to me, there's no 100 million dollar check in New Zealand, right? So you really have to, if you're a great global company, you'll find that money abroad. But, you know, there's extra steps to be able to get that capital. I don't think Europe has the same problems as, you know, a faraway island like New Zealand does, but...

• 22:50Speaker 01

  I think the question of capital is only based on expected return. So if you're addressing only a local market with a cap exit, yeah, it's going to be very hard to raise 100 million because people are not going to see the return on that.

• 23:02Rebecca Bellan?

  Yeah.

• 23:02Speaker 01

  But if you raise 100 million and you can create billions of dollars of value by addressing a global market, then you'll find it wherever you are, right? So it's more a question of what is the potential of the company and the scope and their scope and their willingness and ability to really go global or address a very large market more than the capacity of the venture market to put 100 million in a company.

• 23:28Rebecca Bellan?

  Okay, and before we finish up for the day, do you have any predictions for, I want to say European AI?

• 23:36Speaker 01

  For European AI, that's interesting. So I have, I think that next year is going to be the year of the Agentic Revolution. And I think Europe will play a big part of it. I think we're all very optimistic about H can produce. I mean, it's just a top research team. So, and they have been very focused specifically on that problem. So I've just been looking forward to see what they can do. But I would say in general, next year is when we're going to see agents being able to interact with web application. And basically start to really execute digital tasks for people and for companies. That's going to be super exciting. The other thing that I think is going to happen in the coming year is we're going to start to see

• 24:21Speaker 01

  a new way of developing code because as models are becoming more performance to generate code. I think we're going to move from code completion to application creation. And at that point, the scope of the coding will be to define the specs of the software. And basically developers will write specs and then the AI will write the application and then be able to maintain that application. I mean, that's a vision of a company called Tesol, which is here in London, has been founded by the founder of SNEC, the developer cybersecurity company. And that's exactly the problem that they're working on.

• 25:01Speaker 01

  And I think that's going to be very exciting to see how AI is really going to impact that world.

• 25:07Rebecca Bellan?

  Wow. Okay. Can't wait to see if your predictions pan out. And I think that's it for us today. But where can our listeners connect with you online? Are you online? You on X? You on Threads? You on Mastodon?

• 25:16Speaker 01

  Yeah, so people can find me on my social networks. So I use a lot LinkedIn, which is usually where I put all the content. And I'm also on X.

• 25:25Rebecca Bellan?

  Great. Well, thanks so much for joining us on the show. And to our listeners, you can find me on X at Rebecca Boulon and Threads at Rebecca Boulon underscore writes. And you can always find Equity on X and Threads at Equity Pod. Talk to you Friday.

• 25:38Speaker 01

  Thank you.

• 25:41Speaker 02

  Equity is produced by Teresa Loconsolo with editing by Kel. Bryce Durbin is our illustrator and we'd like to give a big thanks to our audience development team and Henry Pickavet, who manages TechCrunch Audio Products. Thanks so much for listening and we'll talk to you next time.