Generative AI: Revelling in its potential while we grapple with the unknown

We've long wanted to share everything we've been learning at K Fund on the topic of  Generative AI. 

In recent months, we've had fascinating conversations with countless individuals, starting with exchanging opinions amongst ourselves but also with numerous entrepreneurs, colleagues from other funds, academics, researchers, regulators, and people in the public sector. And, of course, with the founders of our portfolio. Moreover, we've spent a significant amount of time diving into the countless articles, threads, podcasts, books, papers, and newsletters currently focused on the space.

Personally, I cannot recall a time quite like this in terms of content generation on a technology-related topic. The volume of information available is, quite simply, overwhelming, making it difficult to discern what is signal and what is noise. From the most generalist newspaper to the most obscure forum in the depths of Reddit, everyone seems to be constantly discussing the virtues and challenges that AI offers us.

When it came to sharing our reflections and our vision (often accompanied by big questions whose answers are still unclear), we thought it would be much more interesting to share the same presentation we've used internally with the entire K team, rather than writing a long article that could become tedious.

In the presentation, you will get to know our vision on Generative AI, how we've built our AI portfolio, the particularities of the Spanish startup scene, and the challenges it faces. It also includes insights from the founders of our portfolio and how tech giants and startups are deploying the technology. It concludes with a reflection on the significant unresolved challenges and a message about the importance of human qualities in this new AI era.

We also thought it was important to talk about the origins of this technology, to highlight key moments in its development, and the most notable advancements up to the turning point that was the launch of ChatGPT last November.

We hope you find it interesting!

Generative AI in Spain

There are approximately 40 GenAI projects in Spain, all focusing on the application layer, leveraging foundational models with a noticeable absence of companies operating in the infrastructure layer.

Half of these projects have fewer than five employees, and 53% have received no funding or have not disclosed their funding status publicly. Whether most of them are proof of concept or ventures with substantial potential is still being determined.

Spanish startups have not announced significant funding rounds despite the rapid escalation of interest in GenAI over recent months. A sense of caution and prudence (and even a bit of scepticism) prevails amongst Spanish VCs, likely contributing to the slower pace of growth and investment. 

Innovation may be hindered by several endemic factors in Spain (and even Europe), including a lack of top-tier AI companies that can act as a breeding ground for future startups, a brain drain of PhDs to foreign corporations, and limited domestic demand for deep tech.

The Spanish GenAI ecosystem is still nascent, with numerous former entrepreneurs interested in returning to the trenches to code or start new companies. Multiple pet projects have also emerged, potentially evolving into serious ventures. 

The founders' profiles are a dynamic blend. They include seasoned entrepreneurs making a comeback or working on side projects, such as Javier Andrés, (ex Ticketea) with LuzIA, or Andrés Torrubia with Lucentum.ai. Makers from various startups are also part of the mix, as seen with Dupla AI (Factorial and Nestor members). Additionally, young coders experimenting with GenAI, like those at Escríbelo, are present.

Projects that have received funding include: Nuclia  (Elaia, Crane. $5.4M Seed round last year), Argilla ($1.6M, Zetta Venture Partners, Caixa Capital Risc), Fermat (Lunar Ventures, undisclosed figures), LuzIA (A* Fund, undisclosed figures), Keytrends (€350k seed round in Mar ‘23 by 4Founders, Draper B1, Tokavi, Ventia Capital), Growthroad (€300k seed round in Feb ‘23 by BAs) and Vidext (#260k 4Founders, BAs).

If you know of any companies we may have missed, please write to me at nacho@kfund.vc. We'll keep updating the map.

Sam Altman in Madrid

Madrid was abuzz with excitement last Monday 22nd for the arrival of Sam Altman, the CEO of OpenAI. Several members of the K team had the opportunity to attend both the fireside chat event and the developers' meeting. The event was hosted by IE University.

Highlighted in the panel discussion was the critical role of regulation in the Generative AI field, focused on both preserving open-source innovation and limiting the power of the LLMs that are emerging at scale. This discourse is set against the backdrop of OpenAI's success, which deviates from the norm by flourishing as a business rooted in a research lab. This success story could serve as a blueprint for Europe's rich research landscape. Parallels were drawn with the early App Store era as we currently navigate through a wave of self-proclaimed AI companies, a phenomenon expected to normalise within a decade. Even for a heavyweight like OpenAI, the hunt for top talent remains a persistent challenge, reinforcing the belief of Altman, in the value of practical experience over formal business education. Looking ahead, he envisions the possibility of $1B valuation startups created by just one or two individuals. This new paradigm could prompt a reassessment of the traditional role of venture capitalists.

We're eager to hear from you

If you're like us, continuous learning on this topic or you are in a journey of building an AI startup, I would absolutely love to hear from you and see how we can help and learn together. 

Acknowledgments

I would like to express my gratitude to my colleagues Max Bray, Miguel Arias, Borja Santos, Jaime Novoa and Jorge Campo for their ideas and unwavering support, which have significantly improved the quality of this report. Thank you so much, mates!

Twitter
Facebook

We've long wanted to share everything we've been learning at K Fund on the topic of  Generative AI. 

In recent months, we've had fascinating conversations with countless individuals, starting with exchanging opinions amongst ourselves but also with numerous entrepreneurs, colleagues from other funds, academics, researchers, regulators, and people in the public sector. And, of course, with the founders of our portfolio. Moreover, we've spent a significant amount of time diving into the countless articles, threads, podcasts, books, papers, and newsletters currently focused on the space.

Personally, I cannot recall a time quite like this in terms of content generation on a technology-related topic. The volume of information available is, quite simply, overwhelming, making it difficult to discern what is signal and what is noise. From the most generalist newspaper to the most obscure forum in the depths of Reddit, everyone seems to be constantly discussing the virtues and challenges that AI offers us.

When it came to sharing our reflections and our vision (often accompanied by big questions whose answers are still unclear), we thought it would be much more interesting to share the same presentation we've used internally with the entire K team, rather than writing a long article that could become tedious.

In the presentation, you will get to know our vision on Generative AI, how we've built our AI portfolio, the particularities of the Spanish startup scene, and the challenges it faces. It also includes insights from the founders of our portfolio and how tech giants and startups are deploying the technology. It concludes with a reflection on the significant unresolved challenges and a message about the importance of human qualities in this new AI era.

We also thought it was important to talk about the origins of this technology, to highlight key moments in its development, and the most notable advancements up to the turning point that was the launch of ChatGPT last November.

We hope you find it interesting!

Generative AI in Spain

There are approximately 40 GenAI projects in Spain, all focusing on the application layer, leveraging foundational models with a noticeable absence of companies operating in the infrastructure layer.

Half of these projects have fewer than five employees, and 53% have received no funding or have not disclosed their funding status publicly. Whether most of them are proof of concept or ventures with substantial potential is still being determined.

Spanish startups have not announced significant funding rounds despite the rapid escalation of interest in GenAI over recent months. A sense of caution and prudence (and even a bit of scepticism) prevails amongst Spanish VCs, likely contributing to the slower pace of growth and investment. 

Innovation may be hindered by several endemic factors in Spain (and even Europe), including a lack of top-tier AI companies that can act as a breeding ground for future startups, a brain drain of PhDs to foreign corporations, and limited domestic demand for deep tech.

The Spanish GenAI ecosystem is still nascent, with numerous former entrepreneurs interested in returning to the trenches to code or start new companies. Multiple pet projects have also emerged, potentially evolving into serious ventures. 

The founders' profiles are a dynamic blend. They include seasoned entrepreneurs making a comeback or working on side projects, such as Javier Andrés, (ex Ticketea) with LuzIA, or Andrés Torrubia with Lucentum.ai. Makers from various startups are also part of the mix, as seen with Dupla AI (Factorial and Nestor members). Additionally, young coders experimenting with GenAI, like those at Escríbelo, are present.

Projects that have received funding include: Nuclia  (Elaia, Crane. $5.4M Seed round last year), Argilla ($1.6M, Zetta Venture Partners, Caixa Capital Risc), Fermat (Lunar Ventures, undisclosed figures), LuzIA (A* Fund, undisclosed figures), Keytrends (€350k seed round in Mar ‘23 by 4Founders, Draper B1, Tokavi, Ventia Capital), Growthroad (€300k seed round in Feb ‘23 by BAs) and Vidext (#260k 4Founders, BAs).

If you know of any companies we may have missed, please write to me at nacho@kfund.vc. We'll keep updating the map.

Sam Altman in Madrid

Madrid was abuzz with excitement last Monday 22nd for the arrival of Sam Altman, the CEO of OpenAI. Several members of the K team had the opportunity to attend both the fireside chat event and the developers' meeting. The event was hosted by IE University.

Highlighted in the panel discussion was the critical role of regulation in the Generative AI field, focused on both preserving open-source innovation and limiting the power of the LLMs that are emerging at scale. This discourse is set against the backdrop of OpenAI's success, which deviates from the norm by flourishing as a business rooted in a research lab. This success story could serve as a blueprint for Europe's rich research landscape. Parallels were drawn with the early App Store era as we currently navigate through a wave of self-proclaimed AI companies, a phenomenon expected to normalise within a decade. Even for a heavyweight like OpenAI, the hunt for top talent remains a persistent challenge, reinforcing the belief of Altman, in the value of practical experience over formal business education. Looking ahead, he envisions the possibility of $1B valuation startups created by just one or two individuals. This new paradigm could prompt a reassessment of the traditional role of venture capitalists.

We're eager to hear from you

If you're like us, continuous learning on this topic or you are in a journey of building an AI startup, I would absolutely love to hear from you and see how we can help and learn together. 

Acknowledgments

I would like to express my gratitude to my colleagues Max Bray, Miguel Arias, Borja Santos, Jaime Novoa and Jorge Campo for their ideas and unwavering support, which have significantly improved the quality of this report. Thank you so much, mates!