Yann LeCun's AMI Labs Raises $1.03 Billion in the Largest European Seed Round Ever — Betting on World Models Over LLMs

A Billion-Dollar Bet Against the LLM Consensus

Yann LeCun has never been shy about criticizing the large language model paradigm — and now he's putting a billion dollars where his mouth is. Advanced Machine Intelligence Labs (AMI), the Paris-based startup LeCun co-founded after departing Meta, announced a $1.03 billion seed round on March 10 at a $3.5 billion pre-money valuation. It's the largest seed round ever for a European startup and one of the biggest AI fundraises of 2026.

The round was co-led by Cathay Innovation, Greycroft, Hiro Capital, and HV Capital, with participation from NVIDIA, Bezos Expeditions, Temasek, Toyota Ventures, Samsung, and Bpifrance Digital Venture. When NVIDIA backs your AI startup alongside Jeff Bezos and three sovereign wealth-adjacent funds, the signal is clear: serious capital believes LeCun's alternative vision has legs.

What Are World Models, and Why Do They Matter?

AMI is built around LeCun's Joint Embedding Predictive Architecture (JEPA), a framework he first proposed in 2022. Unlike autoregressive language models that predict the next token in a sequence, JEPA learns by predicting abstract representations of the world — essentially building an internal model of how things work rather than memorizing patterns in text.

The practical difference is profound. LLMs are fundamentally text-in, text-out systems that simulate understanding through pattern matching. World models, in LeCun's vision, develop genuine causal understanding — they can reason about physics, predict consequences of actions, and plan multi-step strategies in ways that current LLMs cannot. Think of it as the difference between a student who memorized the textbook and one who actually understands the subject.

Why This Matters for the AI Industry

AMI's founding represents the most significant institutional challenge to the LLM-centric approach that dominates the AI industry. OpenAI, Anthropic, Google, and Meta have collectively invested hundreds of billions in scaling language models. LeCun — a Turing Award winner and former chief AI scientist at Meta — is arguing that this entire direction, while commercially successful, is a dead end for achieving truly intelligent systems.

The $1.03 billion war chest gives AMI the resources to test that thesis seriously. The company has already attracted top researchers from Meta FAIR, DeepMind, and INRIA, and its Paris headquarters positions it as a potential European challenger to the American-dominated AI landscape.

Sources: TechCrunch (March 9, 2026), Bloomberg (March 10, 2026), Sifted (March 10, 2026), SiliconANGLE (March 10, 2026)