Yann LeCun Raises $1.03B for AMI, Betting on AI World Models

A Billion-Dollar Bet Against LLMs

In a move that challenges the core trajectory of today's generative AI boom, Turing Award winner and former Meta chief AI scientist Yann LeCun has secured a massive $1.03 billion seed round for his new startup, Advanced Machine Intelligence (AMI). The funding, reported by multiple sources including Reuters and Bloomberg, values the Paris-based company at $3.5 billion pre-money.

AMI's mission is explicit and contrarian: to build AI systems based on "world models" that understand and interact with the physical world. LeCun has long argued that scaling up large language models (LLMs) alone will never achieve 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," he told WIRED.

The AMI Vision: From Engines to Domestic Robots

AMI aims to create AI with persistent memory, reasoning, planning, and safety built in. Its initial focus is on enterprise applications with complex systems. LeCun cited examples like building a realistic world model of an aircraft engine to help manufacturers optimize for efficiency and reliability.

Near-term target customers include manufacturers, automakers, aerospace companies, and biomedical firms. However, the long-term vision is broader. LeCun told Reuters the technology could eventually power consumer applications like domestic robots, which require "some level of common sense to really understand the physical world." He is also in talks with Meta about potentially deploying AMI's models in its Ray-Ban Meta smart glasses.

Star-Studded Founding Team and Global Ambitions

The startup boasts a formidable founding team drawn from LeCun's Meta network and beyond. LeCun serves as Chairman while continuing his professorship at NYU. The CEO is Alexandre LeBrun, former CEO of AI health startup Nabla, which will be AMI's first partner.

Other co-founders include Meta's former VP for Europe, Laurent Solly (COO); former director of research science Michael Rabbat (VP of World Models); former senior director of AI research Pascale Fung (Chief Research & Innovation Officer); and former Google DeepMind researcher Saining Xie (Chief Science Officer).

AMI is global from day one, with planned hubs in Paris (HQ), New York, Montreal, and Singapore. The Singapore office aims to recruit talent and be close to future clients in Asia.

The Funding and the Competitive Landscape

The $1.03 billion round, co-led by investors including Cathay Innovation, Greycroft, Hiro Capital, HV Capital, and Bezos Expeditions, far exceeded initial rumors of a €500 million target. Other notable backers include Mark Cuban, former Google CEO Eric Schmidt, and French billionaire Xavier Niel.

This places AMI in a growing, well-funded niche. Fei-Fei Li's World Labs raised $1 billion last month, and European startup SpAItial secured a large $13 million seed round. AMI Labs CEO Alexandre LeBrun anticipates a trend, telling TechCrunch, "My prediction is that 'world models' will be the next buzzword. In six months, every company will call itself a world model to raise funding."

Philosophical Stance: Open Source and Societal Control

LeCun's departure from Meta was driven by a belief that world model development, with its enterprise applications, was better suited to an independent company that could share development costs with partners. He confirmed to WIRED that while Meta is not an investor, discussions about collaboration are ongoing.

He also staked out a strong philosophical position on AI control. AMI plans to build open-source technology, with LeCun arguing that AI is too powerful to be controlled by any single private company. This aligns him, unusually, with US government concerns about concentrated AI power, as seen in the Pentagon's recent blacklisting of Anthropic.

"I don't think any of us... has any legitimacy to decide for society what is a good or bad use of AI," LeCun stated, emphasizing that democratic processes, not individual CEOs, should set those boundaries.

Why This Matters: A Fundamental Shift in AI's Path

AMI's launch and monumental funding represent more than just another well-capitalized AI startup. It is a direct challenge to the prevailing LLM-centric paradigm championed by OpenAI, Anthropic, Google, and even LeCun's former employer, Meta. LeCun is betting that the path to artificial general intelligence (AGI) runs through a deep, intuitive understanding of physics and cause-and-effect, not just statistical language prediction.

If successful, AMI's technology could unlock new frontiers in robotics, complex system simulation, and scientific discovery. Its focus on enterprise partnerships provides a clear, near-term commercialization path distinct from the consumer-facing chatbot race. The sheer scale of its seed funding demonstrates that heavyweight investors see merit in this alternative approach, potentially signaling the next major battleground in advanced AI research and development.