![](https://d15shllkswkct0.cloudfront.net/wp-content/blogs.dir/1/files/2024/02/unsplash.png)
![](https://d15shllkswkct0.cloudfront.net/wp-content/blogs.dir/1/files/2024/02/unsplash.png)
Bioptimus, a French artificial intelligence startup led by former DeepMind researchers, launched today with $35 million in seed funding.
Healthcare-focused venture capital firm Sofinnova Partners led the investment. It was joined by more than a half-dozen other backers, including Owkin, a well-funded AI startup based in Paris. The latter company is applying machine learning to tasks such as diagnosing patients and speeding up clinical trials.
Bioptimus’ founding chair, Jean-Philippe Vert, is the chief research and development officer of Owkin. He was previously a research lead at Google Brain, a precursor of the search giant’s DeepMind unit. Bioptimus Chief Technology Officer Rodolphe Jenatton also worked at Google’s AI group along with several other members of the founding team.
In addition to raising capital from Owkin as part of the seed round announced today, Bioptimus has inked a technical partnership with the software maker. The alliance will give it access to patient data from several academic hospitals. It said it will use the data to support its product development efforts, which focus on creating what it desires as the “first universal AI foundation model for biology.”
The company has so far shared only high-level information about its technical roadmap. The AI it’s building is intended to help researchers map out “laws of biology that have hitherto remained too complex to be properly understood.” It envisions its software finding use in fields such as biomedicine and environmental science.
“Foundation models in biology are game-changers,” said Sofinnova partner Edward Kliphuis. “They unlock unprecedented potential to personalize medicine, capturing the uniqueness of each individual while harnessing the collective knowledge of all.”
Google’s DeepMind unit, where many of Bioptimus’ founders worked before launching the company, is one of the biggest names in AI-driven biology research. That’s thanks in large part to the AlphaFold deep learning system it introduced in 2018. The system can automatically map out the structure of proteins, a task that had previously eluded computer sciences for decades.
Last year, DeepMind introduced a new version of AlphaFold that can estimate the shape of not only proteins but also several other biological molecules. It also generates more accurate data than the system’s earlier iterations.
Bioptimus said that it will use a compute environment hosted on Amazon Web Services to support its research. AWS offers multiple AI-optimized instances, some of which are powered by internally-developed chips, as well as several services geared specifically towards the healthcare sector. Those services ease tasks such as organizing and analyzing the medical data used in clinical research projects.
The funding round comes three months after Mistral AI, another France-based foundation model developer, closed a $415 million investment. The raise reportedly valued the latter company at around $2 billion. Mistral AI develops open-source large language models with similar capabilities as OpenAI’s ChatGPT.
THANK YOU