Ptechhub
  • News
  • Industries
    • Enterprise IT
    • AI & ML
    • Cybersecurity
    • Finance
    • Telco
  • Brand Hub
    • Lifesight
  • Blogs
No Result
View All Result
  • News
  • Industries
    • Enterprise IT
    • AI & ML
    • Cybersecurity
    • Finance
    • Telco
  • Brand Hub
    • Lifesight
  • Blogs
No Result
View All Result
PtechHub
No Result
View All Result

This AI Model Can Intuit How the Physical World Works

By Wired by By Wired
December 7, 2025
Home AI & ML
Share on FacebookShare on Twitter


The original version of this story appeared in Quanta Magazine.

Here’s a test for infants: Show them a glass of water on a desk. Hide it behind a wooden board. Now move the board toward the glass. If the board keeps going past the glass, as if it weren’t there, are they surprised? Many 6-month-olds are, and by a year, almost all children have an intuitive notion of an object’s permanence, learned through observation. Now some artificial intelligence models do too.

Researchers have developed an AI system that learns about the world via videos and demonstrates a notion of “surprise” when presented with information that goes against the knowledge it has gleaned.

The model, created by Meta and called Video Joint Embedding Predictive Architecture (V-JEPA), does not make any assumptions about the physics of the world contained in the videos. Nonetheless, it can begin to make sense of how the world works.

“Their claims are, a priori, very plausible, and the results are super interesting,” says Micha Heilbron, a cognitive scientist at the University of Amsterdam who studies how brains and artificial systems make sense of the world.

Higher Abstractions

As the engineers who build self-driving cars know, it can be hard to get an AI system to reliably make sense of what it sees. Most systems designed to “understand” videos in order to either classify their content (“a person playing tennis,” for example) or identify the contours of an object—say, a car up ahead—work in what’s called “pixel space.” The model essentially treats every pixel in a video as equal in importance.

But these pixel-space models come with limitations. Imagine trying to make sense of a suburban street. If the scene has cars, traffic lights and trees, the model might focus too much on irrelevant details such as the motion of the leaves. It might miss the color of the traffic light, or the positions of nearby cars. “When you go to images or video, you don’t want to work in [pixel] space because there are too many details you don’t want to model,” said Randall Balestriero, a computer scientist at Brown University.

Yann LeCun, a computer scientist at New York University and the director of AI research at Meta, created JEPA, a predecessor to V-JEPA that works on still images, in 2022.

Photograph: École Polytechnique Université Paris-Saclay



Source link

Tags: Artificial Intelligencequanta magazinescience
By Wired

By Wired

Next Post
Компания Dahua Technology проводит торжественное мероприятие Pro Challenge 2025 для углубления сотрудничества с глобальной экосистемой установщиков

Компания Dahua Technology проводит торжественное мероприятие Pro Challenge 2025 для углубления сотрудничества с глобальной экосистемой установщиков

Recommended.

New wp2shell WordPress Core Flaw Lets Unauthenticated Attackers Run Code

New wp2shell WordPress Core Flaw Lets Unauthenticated Attackers Run Code

July 17, 2026
AI agents help Cato slash ‘time-to-protect’ from new CVEs | Computer Weekly

AI agents help Cato slash ‘time-to-protect’ from new CVEs | Computer Weekly

June 1, 2026

Trending.

Kioxia launches 245TB LC9, the biggest flash drive on the market | Computer Weekly

Kioxia launches 245TB LC9, the biggest flash drive on the market | Computer Weekly

July 28, 2025
OpenAI Really Wants Codex to Shut Up About Goblins

OpenAI Really Wants Codex to Shut Up About Goblins

April 29, 2026
China’s AI Boyfriend Business Is Taking On a Life of Its Own

China’s AI Boyfriend Business Is Taking On a Life of Its Own

January 20, 2026
Google CEO On Gemini 4, Allocating TPUs, AI Models And Gemini Enterprise

Google CEO On Gemini 4, Allocating TPUs, AI Models And Gemini Enterprise

July 28, 2026
Gartner: How AI will transform managed network services | Computer Weekly

Gartner: How AI will transform managed network services | Computer Weekly

May 18, 2026

PTechHub

A tech news platform delivering fresh perspectives, critical insights, and in-depth reporting — beyond the buzz. We cover innovation, policy, and digital culture with clarity, independence, and a sharp editorial edge.

Follow Us

Industries

  • AI & ML
  • Cybersecurity
  • Enterprise IT
  • Finance
  • Telco

Navigation

  • About
  • Advertise
  • Privacy & Policy
  • Contact

Subscribe to Our Newsletter

  • About
  • Advertise
  • Privacy & Policy
  • Contact

Copyright © 2025 | Powered By Porpholio

No Result
View All Result
  • News
  • Industries
    • Enterprise IT
    • AI & ML
    • Cybersecurity
    • Finance
    • Telco
  • Brand Hub
    • Lifesight
  • Blogs

Copyright © 2025 | Powered By Porpholio