Mistral AI raises $113M in seed funding to challenge OpenAI: a promising startup emerges in the race for practical AI

France’s Mistral AI, a four-week-old startup founded by alums from Google’s DeepMind and Meta, has secured a staggering $113 million in seed funding to compete with OpenAI in the development and application of large language models and generative AI. The Paris-based company, led by CEO Arthur Mensch, aims to tackle the current challenge of making AI truly useful by focusing on open-source solutions and targeting enterprise customers. Mistral AI plans to release its first models for text-based generative AI in 2024.

The $113 million seed round was led by Lightspeed Venture Partners, with participation from prominent investors such as Xavier Niel, JCDecaux Holding, Rodolphe Saadé, Motier Ventures, La Famiglia, Headline, Exor Ventures, Sofina, First Minute Capital, and LocalGlobe. Notably, Mistral AI counts French investment bank Bpifrance and former Google CEO Eric Schmidt among its shareholders. The funding values Mistral AI at $260 million, which was the same number that had been rumored in the French press prior to the announcement.

Mensch, along with co-founders Timothée Lacroix (CTO) and Guillaume Lample (Chief Science Officer), have been working together since their school days, all studying artificial intelligence. Mensch previously worked at DeepMind in Paris, while Lacroix and Lample were based at Meta’s Paris AI outpost. They recognized the accelerating pace of AI development, particularly with OpenAI’s GPT model, and saw an opportunity to offer an alternative to the proprietary approach adopted by many companies. Open source is a fundamental part of Mistral AI’s ethos, and the company plans to build models using publicly available data to avoid legal issues related to training data. Users will also be able to contribute their own datasets, and both models and datasets will be open-sourced.

Mistral AI’s unique focus on enterprise customers sets it apart from other players in the AI space, which typically cater to consumer applications. The startup aims to help enterprises understand their AI needs and provide them with easy-to-use tools to create their own AI products. While it is still early to gauge the market reception of Mistral AI’s future products, the company’s emphasis on enterprise customers addresses a perceived gap in the industry.

The sizable funding round for a young company without any customers or products may seem risky to some observers, especially considering recent high-profile cases where startups with Google pedigrees, such as Neeva, pivoted or faced challenges. However, Antoine Moyroud, the investor from Lightspeed Venture Partners, believes the investment is justified. He sees Mistral AI as one of the few teams worldwide with the expertise to build and optimize language models, comparing the AI landscape to infrastructure plays like cloud computing and databases.

The fact that VCs are investing in AI startups indicates that the field is still ripe with opportunities. While major tech companies like Google, Apple, Meta, Amazon, and Microsoft are undoubtedly making their plays in the AI space, the emergence of startups and their ability to secure funding suggests that there is still room for innovation and competition.

Mistral AI intends to utilize the funding to assemble a world-class team that will create the best open-source models. Mensch envisions a collective effort from global players to shape the future of AI and sees Mistral AI as France’s contribution to this endeavor. The name “Mistral” itself, referring to a north wind that brings good weather, symbolizes the startup’s hopes for a positive impact on the AI landscape.