French AI startup Mistral has proposed a new framework for compensating European creative industries: a mandatory levy on AI model revenues. The suggestion comes as debates intensify globally over how to fairly reward artists, writers, and musicians whose work trains AI systems.
The Proposal
Mistral’s leadership has suggested a levy ranging from 1.0 to 1.5 percent of revenues for companies selling AI models in Europe. According to statements from the company, proceeds would flow into a central fund dedicated to supporting cultural industries that have raised concerns about unauthorized use of their work in AI training data.
The proposal positions Mistral—valued at approximately €11.7 billion ($13.5 billion)—as a willing participant in finding market-based solutions to the ongoing tension between AI developers and creative professionals. Unlike some US-based AI companies that have faced litigation over training data, Mistral appears to be taking a more proactive approach to potential regulatory frameworks.
European Context
This proposal emerges against the backdrop of the European Union’s ongoing efforts to establish comprehensive AI regulations through the EU AI Act. France, where Mistral is headquartered, has generally taken a more industry-friendly stance compared to some other EU member states.
The suggestion of alevy mirrors similar mechanisms already in place in other creative sectors. France currently operates a culture levy on blank media and devices, which funds artistic creation—a model that AI policymakers have occasionally referenced when discussing potential compensation frameworks.
Industry Implications
If implemented, such a levy could reshape the economics of AI development in Europe. Smaller players like Mistral might benefit from a more level playing field against well-capitalized US competitors, while the additional costs could potentially be passed on to enterprise customers using European AI services.
The proposal also represents a strategic positioning move for Mistral as the company continues to compete with OpenAI, Anthropic, and Google in the AI race. By proactively engaging with contentious issues around intellectual property and fair compensation, Mistral may be attempting to build goodwill with European regulators and the creative community.
What’s Next
It remains unclear whether European regulators will embrace this specific proposal, or if they might develop alternative frameworks for addressing concerns about AI and creative work. The conversation is expected to continue evolving as more AI companies face pressure to demonstrate how they intend to support—rather than disrupt—the industries they increasingly influence.