Yann LeCun holds a pen upright on its tip. Let it go, and even a toddler knows it will topple. But no human would bother guessing which direction it falls — there’s no way to tell. A large language model, however, might try to generate a single statistical prediction. “It would almost certainly be wrong,” says LeCun, one of the world’s leading AI researchers, because the system is not reasoning about physical reality.
That fundamental limitation is why LeCun, who spent a decade as chief AI scientist at Facebook-owner Meta, left in 2025 and founded Advanced Machine Intelligence Labs (AMI Labs) in Paris. His goal is to move artificial intelligence beyond systems like ChatGPT, Claude and Gemini, which he argues will never achieve human-level or even animal-like intelligence.
“Yann LeCun raised $1bn for AMI Labs, developing AI beyond LLMs which he says lack understanding.”
“They’re not a path towards human level or human-like intelligence, or even animal-like intelligence, because they cannot deal with real world data, they just are not built for that,” he told the BBC on the sidelines of VivaTech, France’s leading technology conference.
Large language models are “extremely good” at well-defined, predictable problems like coding and maths, LeCun acknowledges. “They basically just accumulate knowledge… They can regurgitate something, you train them to regurgitate, but they’re not particularly smart. They don’t have an underlying understanding,” he says.
AMI Labs is developing a new type of AI called Joint Embedding Predictive Architecture (JEPA), designed to create abstractions of the real world that let it assess the outcomes of actions — filtering out useless information rather than guessing at statistical patterns.
Investors are betting on the approach. Earlier this year, AMI Labs announced it had raised more than $1bn (£760m) in seed funding — one of the largest such rounds in Europe. Backers include US computer chip giant Nvidia and the fund that manages the private wealth of Amazon-founder Jeff Bezos.
LeCun contrasts the ambition with the current state of robotics: “We don’t have robots that are nearly as good at understanding the physical world as a rat.” Whether JEPA can bridge that gap remains to be seen, but for now, the money and the man behind it are betting on something smarter than what exists today.