Stay informed with free updates
Simply sign up to the Artificial intelligence myFT Digest — delivered directly to your inbox.
The writer is co-founder and CEO of Mistral AI
Europe is a land of creators. The continent has nurtured ideas that have enriched, and continue to enrich, the world’s intellectual and creative landscape. Its diverse and multilingual heritage remains one of its greatest strengths, central not only to its identity and soft power but also to its economic vitality.
All this is at risk as AI reshapes the global knowledge economy.
Major AI companies in the US and China are developing their models under permissive or non-existent copyright rules, training them domestically on vast amounts of content — including from European sources.
European AI developers, by contrast, operate in a fragmented legal environment that places them at a competitive disadvantage. The current opt-out framework, designed to enable rights holders to protect their content and prevent AI companies from using it for training if they say so, has proven unworkable in practice. Copyrighted works continue to spread uncontrollably online, while the legal mechanisms designed to protect them remain patchy, inconsistently applied and overly complex.
The result is a framework that satisfies no one. Rights holders correctly fear for their livelihoods yet see no clear path to protection. AI developers face legal uncertainty that hampers investment and growth.
Europe needs to explore a new approach.
At Mistral, we are proposing a revenue-based levy that would be applied to all commercial providers placing AI models on the market or putting them into service in Europe, reflecting their use of content publicly available online.
Crucially, this levy would apply equally to providers based abroad, creating a level playing field within the European market and ensuring that foreign AI companies also contribute when they operate here. The proceeds would flow into a central European fund dedicated to investing in new content creation, and supporting Europe’s cultural sectors.
In return, AI developers would gain what they urgently need: legal certainty. The mechanism would shield AI providers from liability for training on materials accessible online. Importantly, it would not replace licensing agreements or the freedom to contract. On the contrary, licensing opportunities should continue to develop and expand for usage beyond training. The fund would complement, not crowd out, direct relationships between creators and AI companies.
We believe in Europe. That is why we are investing €4bn in European infrastructure to train our models on European soil. But we cannot build Europe’s AI future under rules that place us at a structural disadvantage to our US and Chinese competitors. Europe cannot afford to become a passive consumer of technologies designed elsewhere, trained on our knowledge, languages and culture, yet reflecting neither our values nor our diversity.
We are putting forward this idea as a starting point for discussion rather than a final blueprint. With this proposal, we’re inviting creators, rights holders, policymakers and fellow AI developers to come together around a solution where innovation and the protection of creators move forward together.
Europe does not need to choose between protecting its creators and competing in the AI race. It needs a framework that enables both.
The debate around AI and copyright is too often framed as a confrontation between creators and AI developers. This framing is not only unhelpful, it is wrong. Far from being adversaries, the two communities are the most natural of allies. Both have a profound shared interest in ensuring that Europe does not cede ground, culturally, technologically or strategically, in an era that will be defined by how societies choose to govern the tools of intelligence.