Meet Mistral: Europe's Rising AI Star

In the bustling city of Paris, a new star is rising in the world of artificial intelligence. Mistral, a scrappy French startup, is emerging as Europe's best shot at challenging the dominance of tech giants like OpenAI and Google in the fast-moving AI industry.

At the company's airy workspace, a corps of young, sneaker-wearing scientists and programmers tap busily at their keyboards, coding and feeding digital text culled from the internet and reams of 19th-century French literature into Mistral's powerful language model. This model, they believe, will give them the edge they need to take on the tech titans of Silicon Valley.

Leading this charge is Léonard Menga, Mistral's charismatic co-founder and CEO. A brilliant polymath who gravitated toward computers from a young age, Menga has a vision that sets him apart from the tech luminaries across the pond. While figures like Elon Musk and Sam Altman wax poetic about the coming age of artificial general intelligence (AGI) and the potential for machines to surpass human cognition, Menga dismisses such talk as little more than "creating God."

"I'm a strong atheist," Menga declares, his brow furrowed with determination. "I don't believe in God, and I don't believe in AGI."

Instead, the 32-year-old entrepreneur is laser-focused on a more immediate threat: the cultural influence wielded by American AI giants, whose language models are "producing content and shaping our cultural understanding of the world" in ways that don't always align with European values.

This concern has driven Menga and the French government to take a staunchly pro-open-source stance, arguing that allowing others to access and adapt AI models will make the technology safer and more transparent. They've even managed to persuade EU policymakers to limit regulation of open-source AI systems in the bloc's new Artificial Intelligence Act, a victory that Menga believes will help Mistral maintain its rapid development pace.

"If Mistral becomes a big technical power," says Cédric O, the former French digital minister who led the lobbying effort, "it's going to be beneficial for all of Europe."

The Rise of a French 'Genius'

Mistral's origins can be traced back to 2022, when ChatGPT, the groundbreaking language model developed by OpenAI, burst onto the global stage. Menga, who had been working as an academic researcher at France's prestigious National Center for Scientific Research, decided that he and his university friends could do the same or better—right here in France.

"The whole AGI rhetoric is about creating God," Menga says, his eyes narrowing. "I don't believe in God. I'm a strong atheist. So I don't believe in AGI."

Assembling a team of top AI talent, Menga set out to build a more flexible and cost-efficient machine learning tool that could power the kind of chatbots, search functions, and other AI-driven products that captivated the world. And with the full-throated support of the French government, Mistral began to gain ground rapidly.

President Emmanuel Macron has hailed the company as an example of "French genius," Menga has even been invited to dine with the president at the Élysée Palace. Bruno Le Maire, France's finance minister, frequently sings the praises of Mistral, while Cédric O, the former digital minister, has become an adviser to the startup and even owns a stake in the company.

This level of political backing is a testament to the French government's growing recognition of the strategic importance of AI. As countries worldwide race to bolster their domestic AI capabilities, influencing global trade, supply chains, and even foreign policy, Paris has decided that Mistral represents its best chance to challenge the dominance of American and Chinese tech giants.

"This could be one of the best shots that we have in Europe," says Jeannette zu Fürstenberg, the managing director of General Catalyst and a founding partner of the venture capital firm La Famiglia, both of which have invested in Mistral. "You basically have a very potent technology that will unlock value."

Challenging the Tech Titans

Mistral's rapid rise has surprised many in the industry, as the company has developed a machine learning model that can rival the technology coming out of OpenAI. This US startup ignited the AI boom with ChatGPT.

At the heart of Mistral's approach is a commitment to open-source development, a philosophy that sets it apart from the closed-door approach favored by Silicon Valley's tech titans. By making its models available for anyone to download, copy, and adapt, Menga believes Mistral is making its technology safer and more transparent—a direct rebuke to the "fear-mongering lobby" of Google, Microsoft, and Amazon, whom he accuses of trying to "cement their dominance" by persuading policymakers to enact rules that would squash rivals.

"Open source is dangerous, they argue because it has the potential to be co-opted for bad purposes, like spreading disinformation — or even creating destructive AI-powered weapons," Menga says, his voice dripping with sarcasm. "I dismiss such concerns as the narrative of a fear-mongering lobby."

Instead, Menga sees a more immediate threat in the cultural influence wielded by American AI giants, whose language models he believes are "producing content and shaping our cultural understanding of the world" in ways that don't always align with European values.

This concern has prompted Mistral and the French government to push for looser AI regulations in the European Union, arguing that overly restrictive rules would hamper innovation and prevent European companies from competing on the global stage.

"If Mistral becomes a big technical power," says Cédric O, "it's going to be beneficial for all of Europe."

The Challenge Ahead

While Mistral has made impressive strides, the path ahead is not without significant challenges. The company's competitors in the US and China have access to far more resources, with OpenAI having raised a staggering $13 billion and Anthropic securing more than $7.3 billion in funding.

The computing power required to develop cutting-edge AI models is also astronomically expensive, though the French government claims the country's cheap nuclear energy can help meet Mistral's voracious energy demands.

Menga, however, remains undaunted. Driven by a deep-seated belief in the importance of protecting Europe's cultural and political influence, he is convinced that Mistral's open-source approach and focus on transparency will give it the edge it needs to take on the tech titans of Silicon Valley.

"These models are producing content and shaping our cultural understanding of the world," Menga says, his brow furrowed with determination. "And as it turns out, the values of France and the values of the United States differ in subtle but important ways."

As the global AI race intensifies, all eyes are on Mistral and its charismatic leader. Can this scrappy French startup truly challenge the might of American and Chinese tech giants? The future of Europe's digital autonomy may very well hang in the balance.