AI and Sustainability @ Slush

At the end of November 2023, the ST Engineering Ventures team based in London made an inaugural trip to the Slush Startup and Technology Conference, which is Europe’s largest and one of the world’s top 5 startup conferences organised yearly in Helsinki, Finland. Slush originated from Aalto University — it is said that some students went to their dean’s office and asked for money to run an event. They were given some funds, and the rest is history. Today, the event has grown to feature startups from all over Europe and even Asia (we saw representation from Kazakhstan, Japan, Korea, and India). It continues to be run by student volunteers, which has given the event a decidedly “hip vibe”. Event organisers have also taken great pains to ensure sustainability. All food and drinks are vegan (coffee is made with oat milk!), all utensils are recyclable, and recycling bins are everywhere with volunteers standing by to tell you how to dispose of your waste.

The event has a heavy focus on networking — a matchmaking tool, side events, and booth designs focused on allowing people to interact. This is as opposed to other events that are focused on conferences or ‘exhibit-style’ booths. The event space has a meeting area with almost 300 tables purely for people to schedule 30-minute meetings, an area designated for meet-ups where attendees can then find somewhere else to chat, and multiple working areas scattered all over. This means our trip was heavily dominated by scheduled meetings. Over the 2 days of the event, we had over 15 meetings, largely with venture capitals (VCs), incubators, and accelerators.

Some interesting VCs we met:

• Cottonwood Technology Fund, based in New Mexico, U.S. with an office in the Netherlands. Invests only in hard science (does not invest in AI!). New Mexico is an up-and-coming place for hard science, thanks to its good expanse of land for testing and existing science/military laboratories.
• OTB Ventures, a Poland-based VC that started with Central Eastern Europe and is now expanding to the rest of Europe. The co-founders are the founder of Mercuria Energy Group (crude oil trading company), the founder of Expander/Platinum Bank, and a VP of Intel Capital with 18 years of experience — very familiar with corporate VCs.
• Verve Ventures, Zurich-based pan-European VC with offices in Berlin, Paris, Lausanne and Cambridge. Syndicates each investment instead of running a fund and is a highly prolific investor.

Between the meetings we had and the talks we attended, it was clear that Europe has a big focus on AI and sustainability.

In terms of AI, the trajectory that Europe is following is one of Applied AI and the Future of Compute. Europe already has foundational model champions in Stable Diffusion and Mistral, with Mistral stealing the thunder with a highly-attended talk by the CEO, Arthur Mensch. During his talk, Arthur anchored his company as a European champion — not just a French one — as he shared about the importance of Europe having a strong foundational model of its own: “We have a choice of either being on top of the wave and pushing the technology forward or just looking at it happening in the U.S. and China. In order to shape the technology to our values and to the way we think about democracy and society, we need to have very strong technological actors.”

Across the investors we spoke with, no one is interested in investing in foundational models anymore. Everyone is looking for good applications of AI. Interestingly, investors have a common idea of what ‘good’ means — companies should have data that nobody else has or knows how to curate, perhaps from data collected over past projects or extensive operating experience in specific industries. That data should be used to create a strong defensible moat. At the time of the event, a lot of these companies were in the form of developer co-pilots — AI that helps developers code more easily. This includes V0.dev by Vercel, Pandas AI, Poolside, and Diffblue. More recently, we came across a company called Harmonic Security that has built a model to identify when corporate users are about to upload sensitive data to ChatGPT, allowing the employer to have control over what data goes out of the company.

While Europe is no longer interested in foundational models, it remains very much interested in the foundational hardware — the Future of Compute. There is strong interest in semiconductors across the edge and data centre use cases. Edge use cases tend to focus on enabling AI in embedded electronics and would often tout low power and high inference speeds. Meanwhile, data centre use cases may tout low power, but would often focus on higher speeds enabled by new mediums — photonics and quantum. Investors are interested in all of these, thinking of it as a near-term (Edge AI, resistive RAM, spintronics) to long-term (photonics, quantum) timeline.

The topic of sustainability is a much more fragmented one. Depending on the operating experience of the investor, there was a spread of focus across renewable energy, food/AgTech and carbon capture. What was clear, however, is that investors interested in renewable energy would always have an interest in hydrogen as a medium of energy storage. This is being considered as an alternative to batteries for balancing out the ebb and flow nature of green energy production. Meanwhile, conversations around nuclear energy (both fusion and fission) were lacking.

Slush has been a useful eye-opener into the workings of Europe and would be a convenient way to keep an eye on developments outside of the typical regions of the UK, Germany, and France. What is useful to us is that despite the many working languages across Europe, English is the common working language and is what was used at Slush — even the Moldovans and Kazakhs spoke sufficiently fluent English. This seems to be a result of Europe’s strong focus on collaboration and as such, could potentially be a good alternative source of technology for our business units going forward.