Why AI Needs JEPA World Models
Échec de l'ajout au panier.
Échec de l'ajout à la liste d'envies.
Échec de la suppression de la liste d’envies.
Échec du suivi du balado
Ne plus suivre le balado a échoué
Why AI Needs JEPA World Models
-
Narrateur(s):
-
Auteur(s):
À propos de cet audio
The Futurists starts 2026 with a stimulating conversation with serial entrepreneur Matt Miesnieks, a true pioneer of AR/XR and spatial computing. In his new startup venture, Primate AI, Matt is focused on a novel approach to artificial intelligence. He intends to construct spatial and dimensional concepts that replicate the way humans develop a mental model of the real world. Topics in this episode: how the limitations of LLMs create opportunities for new approaches, such as Yann LeCun’s JEPA (Joint Embedding Predictive Architecture); the distinction between trying to understand the real world and trying to generate new worlds; why it is so hard to get a robot to cross a busy street safely; why 3D world models are needed; what happens when the real world is machine-readable.