Who Will Win the Race for AI Supremacy in Europe?

At this point, you certainly don’t have to work in tech to be aware of the world-changing impact of generative AI, whose advent is being widely framed as a new industrial revolution. This is reflected in the feeding frenzy of investment in AI startups across multiple sectors. Those with deep pockets are more than willing to invest in companies in their very earliest stages – see Mistral AI, which secured Europe’s biggest-ever seed round ($113 million) within four weeks of founding.

It's significant that Mistral, hyped as a direct challenger to US-based trailblazer OpenAI, is a French company. That’s because France is vying to become Europe’s capital of all things AI, with the UK arguably its most formidable competitor. That said, other smaller European countries are also going great guns in this burgeoning sphere, so let’s consider some of the factors at play.

The rapid advances being made by firms in the US and China have given European nations a sense of urgency when it comes to accelerating the AI ecosystem on this continent. Given how important AI is becoming to both national economies and national security, it’s no wonder that – as the FT has reported – the policymakers of different countries now “see AI development as a matter of sovereignty.”

In June of these year, French president Emmanuel Macron and British prime minister Rishi Sunak came out with fighting words about AI. Macron pledged €500 million in new funding to create AI “champions”, declaring that France is “number one” when it comes to AI in continental Europe, but that “we have to accelerate”. Meanwhile, Sunak declared his ambition to make the UK the “home of global AI safety regulation”, speaking in the wake of a £1 billion government funding pledge for AI research.

The sense of competition has since become more heated, with rival AI conferences scheduled in the two countries next month. In one corner is a Paris conference backed by French tech billionaire Xavier Niel, who has himself invested €200 million in supercomputer development, saying that “To influence the AI ​​market, you need computing power. To have computing power, you need supercomputers. And to have supercomputers, you have to invest massively.”

In the other corner is the UK’s AI Safety Summit, which takes place a few weeks before the French gathering, and “will bring together international governments, leading AI companies, civil society groups and experts in research.”

According to Sophie Gaston of the UK think tank Policy Exchange, “The British one will be focused on the international diplomacy and governance of AI, while the French proposal seems more industry and innovation focused.”

It’s somewhat ironic that the British conference will be focusing on how best to manage the inherent risks of AI, since many commentators believe the UK will gain a competitive advantage by not coming under the jurisdiction of the AI Act. This is new European Union legislation designed to tackle those inherent risks.

The AI Act establishes strict obligations for providers, depending on the level of risk their technology is deemed to pose. Examples of “high risk” systems include those used in products that fall under EU product safety legislation such as toys, cars and medical devices, as well as AI systems utilised in education and law enforcement. Meanwhile, generative AI tools will have to comply with specific transparency requirements, so disclosures must be made when content is generated by AI.

France will, of course, be bound by this legislation. By contrast, the UK government is taking what it describes as a “pro-innovation approach” to these challenges. It published a white paper earlier this year which bluntly stated how the UK will avoid the kind of “heavy-handed and rigid approach” which could “stifle innovation”. Instead of enacting new legislation, it will implement a “regulator-led approach” which will empower the likes of the Information Commissioner’s Office and Financial Conduct Authority to issue guidance.

According to the UK government, this light-touch approach to regulation will encourage home-grown startups to thrive, and “act as a strong incentive when it comes to AI businesses based overseas establishing a presence in the UK."

Many industry commentators agree that these differing approaches may well give the UK an edge over France when it comes to attracting and nurturing AI tech talent. Speaking to CNBC, Alexandre Lebrun – CEO of healthtech AI startup Nabla – said that, while UK and France are “even” when it comes to providing attractive tech ecosystems, the AI Act could stifle startups to the point where the UK will “win against the EU and France.”

The UK has a headstart in the race for dominance, since it’s home to more AI startups than anywhere else in Europe. However, research by VC firm Earlybird has widened the spotlight well beyond big beasts like Britain and France, highlighting the progress being made in other parts of the continent. 

Estonia, for example, is the leading nation when it comes to number of AI startups per million people. A prominent Estonian success story is the identity verification firm Veriff, which raised an impressive $100 million Series C round last year. The second-place spot, when adjusting for population, is Switzerland, which boasts one of Europe’s most prolific spinout hubs in the form of ETH Zurich (alma mater of Albert Einstein, no less).

These are certainly exciting, disruptive times. As our Sector Principal Heiko Garrelfs puts it, “This is a paradigm shift nobody can afford to miss. The startups that just provide a thin layer of value on top of the OpenAI API were just the first wave.” If you’re a founder or senior decision maker involved in this revolutionary moment, and are looking for expert guidance on a fast-evolving landscape, drop Heiko a line.