AMI Labs, the start-up founded by Yann LeCun, has raised $1.03 billion in a seed funding round that sets a record for European venture capital at that stage, according to reporting by the Financial Times, Reuters, and TechCrunch.
The raise values the company at $3.5 billion, according to the New York Times, a notable entry valuation for a business still at seed stage.
LeCun, who spent more than a decade as chief AI scientist at Meta and is widely credited as one of the founding figures of modern deep learning, left to pursue a technical thesis that diverges sharply from the dominant direction of the industry. Rather than scaling large language models, AMI Labs is focused on building world models: systems designed to develop an internal representation of physical reality, enabling reasoning and planning beyond what text-based models can achieve.
LeCun has been a persistent public critic of the LLM paradigm, arguing that systems trained primarily on text cannot acquire the grounded understanding of the world necessary for general intelligence. AMI Labs represents his attempt to build a practical alternative.
The scale of the raise underscores both the appetite among investors for credible alternatives to the OpenAI-led consensus and the premium placed on founders with LeCun's scientific profile. At $3.5bn post-money, the valuation implies investors are pricing in a long development runway before commercial products emerge.
