Mistral and the dream of becoming a European champion in artificial intelligence

But the road to continental championship status remains an uphill struggle. The first hurdle is one of scale. According to the “State of European Tech 2025” report by Atomico and Dealroom, investment in AI in Europe reached €10.6 billion in 2025 – a record for the continent – but still only a fraction of the €67 billion raised in the United States over the same period.

This gap is reflected in computational resources. Training a state-of-the-art model requires clusters comprising tens of thousands of GPUs and budgets in the region of hundreds of millions of dollars. Mistral has the expertise, but it has to contend with a European financial ecosystem that is still reluctant to back funding rounds worth hundreds of millions.

Then there is the question of talent. Paris has become a magnet – Meta, Google and DeepMind have opened research laboratories in the French capital – but competition for the best researchers is global, and salaries in Silicon Valley remain unbeatable.

Whilst Mistral represents France’s commitment to foundational models, Italia is building a complementary ecosystem made up of smaller start-ups that are focused on high-potential verticals.

iGenius, now renamed Domyn, is the most high-profile example. Founded in Milan ten years ago, it has developed a conversational artificial intelligence platform for analysing business data. Last February, it announced its rebranding and a new European expansion strategy, with the opening of offices in London and Frankfurt. Its stated ambition is to become the continental benchmark for AI applied to business intelligence.

In the field of synthetic data – one of the most promising areas for training AI models whilst respecting privacy – Aindo, a spin-off from the University of Trieste, stands out. The start-up generates artificial datasets that are statistically equivalent to real ones, enabling banks, hospitals and insurance companies to develop algorithms without exposing sensitive information. This puts it in a perfect position for the era of the AI Act.

In the field of customer experience, indigo.ai – a Milan-based conversational AI platform – automates customer service for brands such as Hype, Unobravo, Net Insurance, Flou, Beko and Würth. Its approach involves deep integration with existing business systems and native compliance with European regulations.

Alongside these, a broader ecosystem of more specialised solutions is taking shape. Translated – known for its online translation tool MyMemory – has developed specialised language models for professional machine translation. Userbot combines chatbots and ticketing systems for customer care. VedrAI applies predictive AI to strategic decision-making in SMEs, a sector which struggles to keep pace with technological innovation.

A comparison between France and Italia reveals two different strategies. Paris is banking entirely on a national champion to take to the global stage. Whether by choice or necessity, our country is fostering an ecosystem of vertical specialists – with less capital but greater agility – ready to integrate into European value chains.

Neither approach is a guarantee of success. But together, they paint a more nuanced picture than the adage ‘Europe has no Big Tech’ would suggest.

The game is still wide open. And for once, the Old Continent has a few cards up its sleeve.