A small French AI startup has raised $41 million to develop a biology-based AI model.
As yourself OpenAI's ChatGPT It has taken the world by storm for a super smart next-generation AI device that can speak natural language in text format. Bioptimus Despite taking the concept, training its model for basic biological applications — something that comes with its own unique challenges, as the necessary clinical training data is sensitive and not publicly available.
Bioptimus co-founder and CEO Jean-Philippe Vert The company says it seeks to better understand biology by studying how biology works, from raw data molecules to whole organisms. He said it would allow scientists and researchers to use the biological world to “predict disease outcomes” and develop more effective treatments. It's this simulation that Vert says makes its technology somewhat similar to ChatGPT's underlying model.
“Essentially, it's like biology's GPT, but instead of generating text, we're simulating biology,” he said. Statement.
Bed
France is onto something. A hotbed for AI startupsWith the next generation of AI companies across the country; It secures the lion's share of funding. Same as last year. Mega funding rounds are included. Mistral AI's $640 million coat H insured for $220 million.versus A huggable face It has closed $235 million in investments over the past 18 months.
Bioptimus was founded only last year for its part; But it already has. It raised $35 million in a seed round.. That his foundation has now raised a whopping $76 million in less than a year is a testament not only to the current AI hype, but also to the backgrounds of Bioptimus' six co-founders. Chief Technology Officer (CTO); Rodolphe JenattonFor example, Previously, he was a senior researcher at Amazon and Google. Vert is not only the co-founder and CEO at Bioptimus, but also the Chief R&D Officer. Owkina French unicorn Another thing An AI-infused biotech startup with backers including GV.
These two sections hint at the origins of Bioptimus. Owkin leverages AI and machine learning to accelerate drug discovery and has built several partnerships with leading biopharmaceutical companies. As part of this work, Owkin also collected a large amount of multimodal patient data, which Bioptimus is using to train its underlying model.
It makes more sense to create a separate entity focused on fundamental models than to create a tangential unit within Owkin.
“Building biology (basic models) is not part of Owkin's roadmap, but Owkin is eager to partner with a company like Bioptimus,” Vert explained to TechCrunch in an interview last year. “Course very large (base models) require critical resources for disk volume, computing power and breadth of data models that are easier to unlock as a discrete entity. As a 'pure player' in base models, Bioptimus is better positioned to do this.
In the intervening months, Bioptimus was published. H-Optimus-0 is an open-source foundation model for pathology that has been trained on millions of images to aid in research and diagnosis of diseases such as cancer. However, with a fresh $41 million in the bank, the company will continue to strengthen its AI platform with a more diverse array of data sources covering the pharma and biotech sector and broader therapeutic areas.
As part of this, It is medical, It is preparing to release a new multipurpose foundation model later this year, covering the entire biosphere, which could drive development in sectors including biotechnology and cosmetics.
“Beyond medicine, this model will open limitless possibilities in many other industries and drive biological discoveries in ways we can only imagine,” Vert said.
Bioptimus' latest cash injection was led by US venture capital firm Cathay Innovation and Sofinnova Partners; Bpifrance, Andera Partners; Hitachi Ventures; Boom Capital Ventures; Pomifer Capital, Sunrise and several angel investors are involved.