Advanced Machine Intelligence (AMI), a Paris-based startup co-founded by Yann LeCun, has successfully raised over $1 billion in seed funding to advance its development of AI world models designed to learn and interact with the physical world.This significant funding round represents the largest seed funding for a European startup and positions AMI with a valuation of $3.5 billion.
Sources:
wired.comnews.crunchbase.comLeCun, known for his pivotal role in the AI field, particularly as the former chief AI scientist at Meta, emphasizes the need for AI systems that transcend the limitations of current large language models (LLMs).He argues that human reasoning is fundamentally tied to physical interactions rather than language alone, making the development of world models essential for achieving true human-level intelligence."The idea that you're going to extend the capabilities of LLMs to the point that they're going to have human-level intelligence is complete nonsense," LeCun stated in a recent interview.
Sources:
wired.comobserver.comThe funding was co-led by high-profile investors including Bezos Expeditions, Cathay Innovation, Greycroft, Hiro Capital, and HV Capital, and also included notable backers like Mark Cuban and former Google CEO Eric Schmidt.LeCun plans for AMI to integrate AI applications across various sectors, including healthcare, robotics, and manufacturing, focusing on areas that require complex data analysis and decision-making capabilities.
Sources:
news.crunchbase.comobserver.comAMI's first partnership with Nabla, a healthcare AI startup co-founded by AMI's CEO Alex LeBrun, will provide Nabla with early access to AMI's innovative technologies.This collaboration aims to enhance clinical workflows by developing AI tools that are capable of reasoning and decision-making in real-world healthcare settings.Nabla currently serves over 85,000 clinicians across more than 130 US health systems and has demonstrated significant growth, reporting a five-fold revenue increase in the first half of 2025.
Sources:
statnews.comtheaiinsider.techThe concept of world models proposed by AMI represents a shift from the traditional generative AI approaches that have dominated the market.AMI aims to create systems that not only generate text but also understand and predict outcomes in complex environments.This approach is seen as crucial for applications in healthcare, where understanding intricate workflows and patient interactions can lead to better outcomes and more reliable automation.
Sources:
news.crunchbase.comtheaiinsider.techLeCun's departure from Meta last November was marked by his criticism of LLMs as a “dead end” for achieving genuine intelligence.His vision for AMI focuses on creating AI systems that are controllable, safe, and capable of persistent memory, setting the stage for advancements in various industries that rely heavily on accurate data interpretation and decision-making.
Sources:
observer.comtheaiinsider.techAs AMI moves forward with its plans, the funding and partnerships established will play a crucial role in its mission to redefine how AI interacts with the physical world and to enhance its applicability in vital sectors like healthcare.By creating AI that can reason and plan in a manner akin to human cognition, AMI aims to be at the forefront of a new era in artificial intelligence development, particularly in the context of healthcare, where such capabilities can significantly impact patient care and clinical efficiency.
Sources:
statnews.comwired.comWith ambitious goals and substantial backing, AMI is set to become a key player in the evolving landscape of AI technology, particularly as it relates to health applications.The collaboration with Nabla will likely serve as a stepping stone for further innovations that could revolutionize healthcare and beyond, demonstrating the potential of world model-based AI systems in real-world settings.
