AMI Labs, a new venture founded by Turing Prize winner Yann LeCun after his departure from Meta, has successfully raised $1.03 billion at a pre-money valuation of $3.5 billion. The focus of AMI Labs is on developing world models, which are AI systems that learn from real-world data rather than just textual information.
According to CEO Alexandre LeBrun, the concept of world models is gaining traction in the AI industry, with many companies expected to adopt this approach in the near future. However, he believes that AMI Labs stands out because of its emphasis on understanding the real world, which could have significant applications in sectors like healthcare. The first partnership for AMI Labs is with Nabla, a digital health startup where LeBrun serves as chairman.
LeBrun acknowledges the challenges in transitioning from large language models (LLMs) to world models, especially in ensuring the safety and accuracy of AI systems. AMI Labs is based on the Joint Embedding Predictive Architecture (JEPA) proposed by LeCun, indicating a strong foundation in cutting-edge research.
While the development of world models may take time to reach commercial viability, the sector has already attracted substantial investment. Companies like SpAItial and Fei-Fei Li’s World Labs have secured significant funding, with AMI Labs now joining the ranks with a substantial financial backing.
The success of AMI Labs in fundraising can be attributed to its experienced team, which includes industry veterans like Laurent Solly and prominent researchers like Saining Xie and Pascale Fung. The startup plans to prioritize talent acquisition and research in key locations like Paris, New York, Montreal, and Singapore.
Despite the focus on research and development, AMI Labs aims to engage with potential customers early on to test and refine its world models. The partnership with Nabla is just the beginning, with plans to collaborate with other industry players in the future.
In addition to financial backing from leading investors, AMI Labs has received support from tech giants like NVIDIA and Samsung, as well as French organizations like Groupe Industriel Marcel Dassault and Publicis Groupe. The startup also plans to publish research papers and open-source code to foster a collaborative research ecosystem.
Overall, AMI Labs’ commitment to advancing world models and its collaborative approach to research set it apart in the competitive AI landscape. With a strong foundation in cutting-edge technology and a focus on real-world applications, the startup is poised to make significant contributions to the field of artificial intelligence.
