EQT Ventures is one of Europe’s largest VC investors, raising €2.6 billion across three funds and investing from Seed to Series C. Together, EQT Group has invested in more than 300 companies, including unicorns such as Einride, Wolt, Marvel Fusion and Xeltis, as well as Freepik and Beamery.
To understand how we approach the evolving landscape of AI, defensibility, and early-stage investing, I spoke to Alexander Fred-Ojala, Head of AI at EQT Ventures.
Fred-Ojala leads the AI team that develops the platforms, tools, and algorithms that make EQT’s early-stage investments AI and data-driven. We also advise on the technical aspects of the investment process and help portfolio companies enhance their AI capabilities.
Inside EQT’s Mother Brain: The AI Engine Rewiring Venture Procurement
When it comes to AI tools, EQT is perhaps best known for its proprietary platform Motherbrain, which the company describes as a “powerful synergy of AI and human expertise.”
First launched by EQT Ventures in 2016 as a way to evaluate a large number of technology startups, Motherbrain uses AI to scan, model and track investment opportunities, supporting the entire investment lifecycle from sourcing through due diligence to portfolio value creation.
For early stage investments, there are literally millions of opportunities to evaluate whether you are active in Europe, the US, or around the world. However, humans cannot analyze all these data points. “Mother Brain has been in operation for over 10 years,” Fred Ojala elaborates.
“What we’re trying to do with these internal tools is build systems that give us an edge and increase productivity. For venture companies, a data-driven investment approach can generate alpha, especially in sourcing and finding new opportunities.”
Historically, this has been accomplished through statistical methods such as monitoring website traffic and using predictive models to predict whether a company is on a good trajectory.
“There were so many conversations internally about the opportunity, including thoughts about the founders, market notes, and past valuations,” he said.
Previously, someone had to record and read all of these notes to remember the context. Now, LLMs can analyze that and proactively say, “I don’t know.
“This team member has met the company before, knows the founders, and the discussion was about X. This is the perfect angle to reconnect,” Fred-Ojala shared.
According to Fred-Ojala, what has changed with generative AI and large-scale language models since GPT-3 (2020), and even more since ChatGPT, is that they can now analyze unstructured data. The model can understand all text data online, including company conversations, PDFs, market research, academic papers, and even internal data.
On the procurement side, the company has built an operating system to navigate opportunities. Deal makers log in and see all related opportunities for which digital traces exist. These are stack-ranked using an interpretable scoring model. Dealers can see why something is ranked.
“For example, team members can perform detailed agent evaluations on deals. The system analyzes all publicly available data on deals and cross-runs it with internal data. The main bottleneck to doing this at scale is compute.
The tool also integrates metadata such as “which EQT people in the company know about this company” and “which advisors have met them” to aid in network intelligence and leverage the scale of the EQT platform.
They are also ranked by traction, base, and relevance to EQT Ventures. “You’re less likely to invest in a 20-year-old company.”
“We are looking for early-stage companies with high potential.”
“People still underestimate how transformative AI will be.”
I asked Fred O’Hara whether he saw Europe as a fundamental model region, or whether the greatest opportunities lay elsewhere. He acknowledged that while his role is focused on building internal AI systems, he also works closely with deal teams as the firm evaluates AI opportunities.
He argues that Europe has leaders when it comes to developing frontier models.
“Mistral is a clear example of that. They are only a few months behind the cutting edge, maybe three to six months, but they are innovating in a way that directly competes with the big labs.
And don’t forget, DeepMind was founded in the UK and still powers most of Google’s AI advances. ”
However, building a frontier model is very capital-intensive and Europe may not need to compete. Instead, he suggests:
“We may actually be at the ‘peak size of the underlying model,’ and I wouldn’t be surprised if the market corrects. Short-term hype or bubble? Maybe so. But in the long term, I think people are still underestimating how transformative AI will be.”
However, Fred-Ojala and I share the opinion that Europe is the strongest in AI applications (at least for now). In Europe, companies such as Lovable, Cradle, Leya, and Helsing are building industry-specific vertical solutions based on foundational models.
He shared:
“Europe can accelerate here because our markets are so diverse and nuanced. AI thrives in environments where it can optimize complex workflows, and Europe is full of AI.
I believe that Europe’s greatest opportunities lie there. ”
Agentic AI rewrites every knowledge field
Fred Ojala has no doubts that we are in more than just an AI bubble, but perhaps the biggest paradigm shift in the history of technology. He says this is comparable to electricity and the internet.
“We’re developing an external cognitive engine that solves problems. That changes everything.”
He believes white-collar jobs and digital workflows are the beginning, arguing:
“In a few years, typing on a keyboard and staring at a static screen will feel outdated. Instead, we will be directing systems to manage agent workflows.”
Look at software engineering. OpenAI just launched an agent creation framework. We built the prototype in-house in six weeks, and our proprietary code generation agent, Codex, created 80% of the code.
Two years ago, that would have been science fiction. ”
In the future, agent work will span all fields of knowledge, including biotechnology, chemistry, physics, and mathematics.
“GPT-5 Pro solves new math problems. DeepMind’s AlphaEvolve discovers new algorithms. Goosebumps-inducing moments.”
Regarding bubbles, he argues that short-term bubbles are more related to capital investment (data center expansion) than venture capital itself.
Crypto looked for trouble. AI will solve them
Like many of us, Fred Ojala has lived through the blockchain hype cycle of the ICO era, Web3, NFTs, and business models that rely on platform economics.
Prior to joining EQT Ventures, Fred-Ojala was Research Director of the Data Lab at the Starja Center for Entrepreneurship and Technology (SCET) at the University of California, Berkeley, where he co-founded the Berkeley Blockchain Xcelerator. Its alumni have collectively raised more than $600 million.
He admits that cryptocurrencies often felt like techies looking for a problem. AI is the opposite.
“This solves the problem. Even if progress stopped today, there would be tremendous value created by distributing and integrating current capabilities.”
The main bottleneck is not technology, but people. Changing your workflow, mindset, and habits takes time. This is why enterprise ROI measurements often seem sloppy. ”
Rise of a new power center
As our conversation moved from the hype cycle to real-world implications, Fred Ojala pointed out that the next wave of AI value will not be limited to familiar hubs like London, Paris, and Berlin.
Fred-Ojala explained that EQT recently analyzed AI activity across Europe and found that innovation is occurring far beyond the usual regional power centres.
In turn, AI accelerates research in physics, chemistry, materials, and optimization, opening new physical-world applications and durable business opportunities.
Additionally, “we ended up creating three categories,” he explained. The first are full-stack powerhouse regions, meaning regions with a large number of AI startups and large amounts of investment. “Stockholm stands out here, with more than 50 AI startups and more than 205 million euros invested,” he pointed out.
The second group, Founder Factories, consists of cities with high startup density but relatively little funding. “Tallinn actually has the highest number of AI startups per capita in Europe, with 360 companies per million residents,” he said.
Finally, there is the money magnet, an ecosystem with few startups but high capital inflows. “Heidelberg and Cambridge fall into this category,” explained Fred O’Hara.
“These are deep tech academic cities that always collect big checks.”
Complexity is no longer a moat
When asked how he differentiates between defensibility and hype in today’s AI market, Fred Ojala was clear:
“One of the big changes is that technical complexity is no longer the moat, especially in software,” he said.
Tools like Lovable “allow you to quickly prototype SaaS products like Slack and let engineers refine the parts that AI can’t complete,” he said.
Rather, the attributes that make AI startups truly fundable are changing.
He argues that speed is key, saying, “The teams that adopt AI workflows the fastest will outperform the others.
“Distribution and brand are important assets because mindshare is growing globally and deep domain expertise is now a key differentiator, the kind of vertical focus that Frontier Labs cannot easily replace.”
At the end of the day, his advice to founders is:
“Solve one or two problems very well. Don’t try to solve 50 problems. One-size-fits-all wrappers don’t last.”
“We’ve only scratched the surface.”
When asked where the AI landscape is heading in the next three to five years, Fred Ojala cautioned that while exponential progress is always difficult to predict, the direction is clear.
“The service industry is going to be significantly disrupted,” he said, pointing to a future where end-to-end GenAI workflows take over much of customer support, legal, finance and accounting.
He also predicts that software itself will become increasingly fluid.
“We will see interfaces generated in real time, software that reshapes itself to suit the user.”
Fred Ojala believes some long-dormant technologies are poised for a comeback.
“AR and VR are coming back,” he said, adding that humanoid robotics is finally nearing a level of maturity where it can become a “meaningful contributor” to the overall industry.
He stressed that even if the pace of AI progress slows, we are still in the early stages.
“We have only scratched the surface of current capabilities.”
