OpenAI, Anthropic Pressure: Can European AI Startups Compete?

Artificial intelligence firms are the darlings of today’s startup landscape, yet the breakneck pace is largely dictated by advancements from industry titans like OpenAI and Anthropic. For startups architecting solutions atop these models, the challenge is clear: adapt or become obsolete.

While the U.S. currently leads in the large language model (LLM) arena, an arms race fueled by massive capital investments, Europe’s potential lies in crafting practical AI tools – the application layer. This strategic focus could unlock significant value.

“That’s also where we think most of the profit will be made in the future,” Robert Lacher, a founding partner of Visionaries Club, noted earlier this year.

Generative AI ventures globally secured a staggering $49.2 billion in venture capital (VC) funding in the first half of 2025, surpassing 2024’s $44.2 billion total, according to consultancy EY. However, the U.S. dominates in terms of deal value, accounting for 97%, and 62% of volume. Europe lags behind, representing just 2% of value, yet holds a respectable 23% of volume.

A confluence of factors contributes to this disparity. VC investors in Europe typically exhibit a more conservative risk appetite than their U.S. counterparts. Market fragmentation has historically impeded the rapid scaling of European startups. Furthermore, amidst economic headwinds and a recalibration following the 2021 tech boom, investors are increasingly prioritizing sustainable growth and robust business fundamentals. While AI maintains investor interest, it does not garner the same fervor as in the U.S.

The rapid iteration of foundational AI models like OpenAI’s ChatGPT and Anthropic’s Claude demands that application-layer companies accelerate their development cycles to remain competitive.

Europe has a contender in the LLM space: Mistral, the French startup that has amassed 1.7 billion euros ($2 billion) in funding, with support from Dutch chipmaker ASML. Positioned as an open-source alternative to OpenAI, Mistral faces a steep climb.

“The speed of innovation, speed of product velocity, speed of distribution, actually ends up winning over everything else,” Bryan Kim, a partner at VC firm Andreessen Horowitz, stated recently from Italian Tech Week.

Companies like Sweden’s Lovable, an AI-powered “vibe-coding” platform, and AI agent startup Sana exemplify how AI is being leveraged for practical applications. London’s Synthesia, specializing in AI video generation, and ElevenLabs, focused on synthetic audio, are also building specialized AI applications. Notably, ElevenLabs has since developed its own LLM, highlighting the trend of application-layer companies investing in foundational models.

“How do you move any slower than changes in the underlying product and technology and still expect to win?” Kim questioned, highlighting the pressures facing companies built on rapidly evolving models.

“Momentum is the moat at this current juncture of AI development. Maybe we’ll get to a point where the model layer stabilizes it a little bit, and then we could talk about other things, but, right now, momentum is the only moat that I see,” he added. This suggests that continuous iteration and rapid adaptation are critical for long-term survival in the current AI landscape.

Maintaining momentum – particularly the ability to constantly iterate – is predicated on securing capital for scaling operations.

“If you look at the Europeans, we are revolutionary, we are romantics, we are resourceful,” Jean La Rochebrochard, managing director at Kima Ventures, observed. However, “it’s hard to compete with a country where the appetite for risk is way higher, where the amount of capital is way higher as well, and the talent,” he said, speaking about the US and the broader AI landscape.

La Rochebrochard is still optimistic that Europe can cultivate the next major success story. He believes founders who have gained experience outside of Europe and then return to launch a new venture are particularly promising.

“We do all hope that Mistral will become one of these behemoths, one of these $100 billion companies in Europe, just like Revolut did in the UK. If Revolut, Mistral and Spotify are doing it, why not another 10, 20, 50 others?” the investor added. This emphasizes the importance of building upon existing success stories to foster a more robust ecosystem for AI innovation in Europe.

The UK-based AI cloud firm Nscale recently secured $433 million in new funding, following a $1.1 billion Series B – the largest in Europe – just days prior. Like Mistral, Nscale’s focus is AI infrastructure rather than the application layer. This development reflects the increasing significance of AI sovereignty for policymakers and investors. Building independent AI infrastructure is seen as crucial to reducing reliance on U.S. technology and fostering greater control over data and algorithms.

However, for Lovable CEO Anton Osika, the core issue is attitudinal: “The only thing we need to do in Europe is change our mindset that it is possible.”

“Traditionally it has been more of a constraint with access to the amount of technical talent, of access to capital, that is not the bottleneck anymore,” he argued. The availability of talent and capital is no longer the primary impediment, highlighting the importance of fostering a more ambitious and confident mindset among European entrepreneurs.

Osika argues that European AI companies now possess a competitive advantage in talent acquisition: “It’s much faster for us to hire in Europe than it is to do so for U.S. counterparts, where there’s 1,000 more companies like Lovable, so it is a competitive advantage to be building from Europe.” This points to the potential for Europe to attract and retain top AI talent, further bolstering its position in the global AI landscape.