“We do not have robots which can be almost nearly as good at understanding the bodily world as a rat,” says Yann LeCun, one of many main figures on the planet of synthetic intelligence.
He labored at Fb-owner, Meta, for a decade, the place he was chief AI scientist, however left in 2025 and based Superior Machine Intelligence Labs (AMI Labs).
His objective is to maneuver AI past present programs like ChatGPT, Claude and Gemini. They’ve their makes use of, he says, however won’t ever have the ability to deal with difficult conditions in the actual world, like getting a robotic to do family chores.
“They are not a path in direction of human degree or human-like intelligence, and even animal-like intelligence, as a result of they can’t cope with actual world information, they simply should not constructed for that,” he tells me on the sidelines of VivaTech, France’s main know-how convention.
So, Paris-based AMI Labs is busy creating a brand new sort of synthetic intelligence not primarily based on the tech behind ChatGPT and its rivals.
Buyers assume it has potential. Earlier this 12 months AMI Labs introduced that it had raised greater than $1bn (£760m), with traders together with US pc chip large Nvidia and the fund that manages the non-public wealth of Amazon-founder Jeff Bezos.
That so-called seed funding spherical – the earliest spherical of start-up fundraising – was one of many greatest of its variety in Europe.
Giant Language Fashions (LLMs) like ChatGPT are extraordinarily good at some issues like coding, mathematical issues and producing textual content, LeCun says.
However he argues that these are properly outlined and predictable issues.
“They [LLMs] mainly simply accumulate data… They’ll regurgitate one thing, you prepare them to regurgitate, however they are not significantly sensible. They do not have an underlying understanding,” he says.
In the actual world there’s a bewildering array of outcomes to any motion, which requires a extra versatile sort of synthetic intelligence.
LeCun holds a pen upright on its tip. What occurs if you let go, he asks? Even a toddler would know that the pen would topple over. However no human would hassle to guess by which route the pen would possibly fall, there isn’t any solution to inform.
However an LLM would possibly attempt to generate a single prediction concerning the pen’s subsequent transfer primarily based on statistical patterns from its coaching information.
The prediction would virtually definitely be unsuitable, as a result of the system just isn’t reasoning concerning the bodily actuality of the state of affairs – it’s producing what seems to be statistically believable.
LeCun says the system his firm is creating, known as Joint Embedding Predictive Structure (JEPA), is ready as much as cope with issues like that.
It creates abstractions of the actual world that permit it to evaluate the outcomes of actions.
Creating these abstractions includes troublesome maths, however basically they filter out ineffective info, simply leaving the AI with helpful photos of the world.
Within the case of the pen, the AI would know that there isn’t any level in attempting to foretell which method the pen would fall.
