Superior Machine Intelligence (AMI), a brand new Paris-based startup cofounded by Meta’s former chief AI scientist Yann LeCun, introduced Monday it has raised greater than $1 billion to develop AI world models.
LeCun argues that almost all human reasoning is grounded within the bodily world, not language, and that AI world fashions are essential to develop true human-level intelligence. “The concept that you’re going to increase the capabilities of LLMs [large language models] to the purpose that they’re going to have human-level intelligence is full nonsense,” he mentioned in an interview with WIRED.
The financing, which values the startup at $3.5 billion, was co-led by buyers corresponding to Cathay Innovation, Greycroft, Hiro Capital, HV Capital, and Bezos Expeditions. Different notable backers embody Mark Cuban, former Google CEO Eric Schmidt, and French billionaire and telecommunications govt Xavier Niel.
AMI (pronounced just like the French phrase for pal) goals to construct “a brand new breed of AI techniques that perceive the world, have persistent reminiscence, can motive and plan, and are controllable and protected,” the corporate says in a press launch. The startup says will probably be international from day one, with places of work in Paris, Montreal, Singapore, and New York, the place LeCun will proceed working as a New York College professor along with main the startup. AMI would be the first business endeavor for LeCun since his departure from Meta in November 2025.
LeCun’s startup represents a wager in opposition to most of the world’s greatest AI labs like OpenAI, Anthropic, and even his former office, Meta, which imagine that scaling up LLMs will ultimately ship AI techniques with human-level intelligence and even superintelligence. LLMs have powered viral merchandise corresponding to ChatGPT and Claude Code, however LeCun has been one of many AI business’s most outstanding researchers talking out concerning the limitations of those AI fashions. LeCun is well-known for being outspoken, however as a pioneer of contemporary AI that received a Turing award again in 2018, his skepticism carries weight.
LeCun says AMI goals to work with corporations in manufacturing, biomedical, robotics, and different industries which have plenty of information. For instance, he says AMI may construct a practical world mannequin of an plane engine and work with the producer to assist them optimize for effectivity, reduce emissions, or guarantee reliability.
AMI was cofounded by LeCun and a number of other leaders he labored with at Meta, together with the corporate’s former director of analysis science, Michael Rabbat; former vice chairman of Europe, Laurent Solly; and former senior director of AI analysis, Pascale Fung. Different cofounders embody Alexandre LeBrun, former CEO of the AI well being care startup Nabla, who will function AMI’s CEO, and Saining Xie, a former Google DeepMind researcher who would be the startup’s chief science officer.
The Case for World Fashions
LeCun doesn’t dismiss the general utility of LLMs. Relatively, in his view, these AI fashions are merely the tech business’s newest promising pattern, and their success has created a “form of delusion” among the many individuals who construct them. “It is true that [LLMs] have gotten actually good at producing code, and it is true that they’re most likely going to change into much more helpful in a large space of purposes the place code technology might help,” says LeCun. “That’s quite a lot of purposes, however it’s not going to result in human-level intelligence in any respect.”
LeCun has been engaged on world fashions for years inside Meta, the place he based the corporate’s Elementary AI Analysis lab, FAIR. However he’s now satisfied his analysis is finest achieved exterior the social media big. He says it’s change into clear to him that the strongest purposes of world fashions might be promoting them to different enterprises, which doesn’t match neatly into Meta’s core client enterprise.
As AI world fashions like Meta’s Joint-Embedding Predictive Architecture (JEPA) grew to become extra subtle, “there was a reorientation of Meta’s technique the place it needed to principally meet up with the business on LLMs and form of do the identical factor that different LLM corporations are doing, which isn’t my curiosity,” says LeCun. “So someday in November, I went to see Mark Zuckerberg and informed him. He’s at all times been very supportive of [world model research], however I informed him I can do that sooner, cheaper, and higher exterior of Meta. I can share the price of improvement with different corporations … His reply was, OK, we will work collectively.”
