AMI Labs, the new venture cofounded by Turing Prize winner Yann LeCun after his departure from Meta, has closed a $1.03 billion funding round at a $3.5 billion pre-money valuation. The lab’s mission: build “world models” — AI systems that learn about the real world from multi-modal, real-world data rather than relying solely on language.
What are world models and why they matter 🤖🌍
World models aim to create AI that understands environments, events, and physical realities — not just patterns in text. That difference could be critical for high-stakes fields like healthcare, robotics, and scientific discovery, where hallucinations from large language models (LLMs) pose serious risks.
AMI Labs will pursue research grounded in architectures like JEPA (Joint Embedding Predictive Architecture), proposed by LeCun in 2022, with the goal of moving from theoretical foundations to robust, real-world evaluations.
Big funding, long horizon 💰🧪
Unlike many applied AI startups chasing quick product launches and near-term revenue, AMI Labs is positioning itself as a deep research lab. CEO Alexandre LeBrun emphasizes that this is “a very ambitious project, because it starts with fundamental research. It’s not your typical applied AI startup that can release a product in three months.” In short: the path from lab breakthroughs to commercial deployments may take years.
Why investors backed a long-term bet
Despite the extended timeline, world-model efforts have attracted large checks recently — from Europe’s SpAItial raising a $13 million seed to Fei-Fei Li’s World Labs securing roughly $1 billion. AMI Labs’ $1.03 billion round was reportedly oversubscribed: the team initially sought around €500 million last December but closed closer to €890 million — likely reflecting investor confidence in the founding team and strategy.
Leadership, partners, and where the work will happen 🧑🔬
AMI Labs’ leadership blends research credibility and operational experience:
- Yann LeCun — chairman and cofounder, Turing Prize winner;
- Alexandre LeBrun — CEO, entrepreneur and former CEO of Nabla;
- Laurent Solly — COO and Meta’s former VP for Europe;
- Saining Xie — chief science officer;
- Pascale Fung — chief research & innovation officer;
- Michael Rabbat — VP of world models.
AMI Labs will build teams in four hubs: Paris (headquarters), New York (LeCun’s NYU base), Montreal (where Rabbat is located), and Singapore to recruit talent and engage Asia-based partners.
Early partnerships and real-world testing 🔬
Although AMI Labs does not plan to generate revenue in the near term, it intends to work with partners early to validate models in real settings. Nabla, a digital health startup where LeBrun serves as chairman, is the first disclosed partner expected to access early models — a natural fit given healthcare’s need for trustworthy, grounded AI.
LeBrun notes: “We are developing world models that seek to understand the world, and you can’t do that locked up in a lab. At some point, we need to put the model in a real-world situation with real data and real evaluations.”
Who invested — and what they bring
The round was co-led by Cathay Innovation, Greycroft, Hiro Capital, HV Capital and Bezos Expeditions, with participation from institutional and strategic backers including NVIDIA, Samsung, Sea, Temasek and Toyota Ventures. Notable individual investors include Tim and Rosemary Berners-Lee, Jim Breyer, Mark Cuban, Mark Leslie, Xavier Niel and Eric Schmidt. French industrial and investment groups such as Association Familiale Mulliez, Groupe Industriel Marcel Dassault and Publicis Groupe also participated.
LeBrun says the strong investor interest allowed AMI Labs to select partners aligned with its long-term scientific and ethical expectations, not just capital.
Compute, talent and an open research ethos ⚙️📚
Major cost centers for AMI Labs will be compute and top-tier talent. The funding gives the lab runway to hire selectively and secure the compute resources needed for large multi-modal experiments. Rather than scale quickly for revenue, the priority is building a high-quality team and infrastructure.
AMI Labs also plans to publish research and open-source significant amounts of code. LeBrun argues that open science accelerates progress: “We think things move faster when they’re open, and it’s in our best interest to build a community and a research ecosystem around us.”
Bottom line
AMI Labs’ $1.03 billion raise signals that investors are willing to back ambitious, long-term bets on AI that understands the physical world. The lab’s combination of top researchers, deep pockets, and an open approach positions it as a major player in the emerging field of world models — but substantial work remains to turn theory into robust, safe, and deployable systems. 🚀
Key takeaways: AMI Labs raised $1.03B, will focus on JEPA-style world models, partners early with companies like Nabla, plans open research, and expects a multi-year path to real-world applications.