Connect with us

AI That Connects the Digital and Physical Worlds | Anima Anandkumar | TED

“While language models may help generate new ideas, they cannot attack the hard part of science, which is simulating the necessary physics,” says AI professor Anima Anandkumar. She explains how her team developed neural operators — AI trained on the finest details of the real world — to bridge this gap, sharing recent projects ranging…

Published

on

“While language models may help generate new ideas, they cannot attack the hard part of science, which is simulating the necessary physics,” says AI professor Anima Anandkumar. She explains how her team developed neural operators — AI trained on the finest details of the real world — to bridge this gap, sharing recent projects ranging from improved weather forecasting to cutting-edge medical device design that demonstrate the power of AI with universal physical understanding.

If you love watching TED Talks like this one, become a TED Member to support our mission of spreading ideas:

Follow TED!
X:
Instagram:
Facebook:
LinkedIn:
TikTok:

The TED Talks channel features talks, performances and original series from the world’s leading thinkers and doers. Subscribe to our channel for videos on Technology, Entertainment and Design — plus science, business, global issues, the arts and more. Visit to get our entire library of TED Talks, transcripts, translations, personalized talk recommendations and more.

Watch more:

TED’s videos may be used for non-commercial purposes under a Creative Commons License, Attribution–Non Commercial–No Derivatives (or the CC BY – NC – ND 4.0 International) and in accordance with our TED Talks Usage Policy: . For more information on using TED for commercial purposes (e.g. employee learning, in a film or online course), please submit a Media Request at

#TED #TEDTalks

Continue Reading
Advertisement
35 Comments

35 Comments

  1. @MichaelSuperbacker

    July 15, 2024 at 7:00 am

    Hello ????

  2. @lyrics4daheart

    July 15, 2024 at 7:03 am

    Thank you

  3. @user-vp4ss6wk9h

    July 15, 2024 at 7:07 am

    3rd

  4. @MatrixEvolution17

    July 15, 2024 at 7:11 am

    Why are there so many AI videos now? NO ONE CARES.

    • @sparkofcuriousity

      July 16, 2024 at 6:07 am

      hahaha

  5. @MC-br1gk

    July 15, 2024 at 7:22 am

    Here are these smiling pretty faces pushing tech on us again. I’m so disgusted by the damaging tech industry and their proponents. We are pushing tech so far from our cultural skillset to deal with it. F TED.

  6. @jeffbezozss

    July 15, 2024 at 7:23 am

    The video smells

  7. @imdxpertm

    July 15, 2024 at 7:26 am

    Amazing work ! This talk gives a very impressive future that awaits when AI is put to use for betterment of humanity. Wishing you all the best!

  8. @Youtube_Summarizer

    July 15, 2024 at 7:31 am

    Here are the key points and the important details of this video, created by @Youtube_Summarizer

    Key Points:
    1. [Key Point 1]: AI transforms science, engineering with physical, digital integration.
    2. [Key Point 2]: Neural operators enable faster, accurate simulations of complex phenomena.
    3. [Key Point 3]: AI-based weather model outperforms traditional models in speed, accuracy.
    Important Details:
    Here’s the timeline, created by @Youtube_Summarizer
    00:00:04 Inspired by Parents’ Work in Manufacturing
    • Speaker was inspired by parents’ work in computerized manufacturing.
    • Fascinated by how computer programs affected the physical world.
    00:00:29 AI’s Role in Science and Engineering
    • AI aims to connect physical and digital worlds.
    • Reduces trial and error in scientific research and engineering design.
    00:01:22 Challenges of Language Models
    • Language models lack physical grounding, hallucinate.
    • AI needs data from the world to understand physics.
    00:03:01 Neural Operators for Detailed Simulations
    • Neural operators represent data as continuous functions.
    • Allows infinite zoom and multi-scale learning.
    00:04:03 Application in Medical Catheter Design
    • Neural operators helped design a better medical catheter.
    • Reduced bacterial contamination by over 100-fold.
    00:05:59 AI’s Potential in Weather Forecasting
    • AI created a faster, more accurate weather model.
    • FourCastNet model outperformed traditional models in accuracy.
    00:07:57 AI’s Role in Nuclear Fusion
    • AI predicts plasma evolution in nuclear fusion reactors.
    • Enables real-world corrective actions for fusion reactors.
    00:08:55 Vision for Generalist AI
    • Aiming to create an AI model for all scientific problems.
    • Generalist AI will simulate any physical phenomena and design novel solutions.

    We hope this summary helps you better understand the video content! If you have other videos that need to be summarized, please visit @Youtube_Summarizer

  9. @liewziqin2319

    July 15, 2024 at 7:42 am

    Just curious why can’t we run some simulations to verify before printing the design? I agree that AI can propose the designs but I think we should carefully simulate the process to a certain accuracy before printing it out? Especially for something expensive to build ????

    • @jeevan88888

      July 15, 2024 at 1:48 pm

      before printing out what?

    • @ernestodanieljaramillo9522

      July 17, 2024 at 12:35 am

      ​@@jeevan88888 the designs or draft of any solution against certains problems

  10. @Ken-yp1dg

    July 15, 2024 at 7:43 am

    AI doesn’t exist yet. Companies wanted to use the term for marketing.

    • @user-zq1fn9vo1x

      July 15, 2024 at 9:34 am

      wdym

    • @WeylandLabs

      July 15, 2024 at 10:57 am

      Wrong LLM’s are a form of A.I please educate yourself more sir.

  11. @antonioas709

    July 15, 2024 at 7:54 am

    Right this won’t be used for the wrong reasons

  12. @rw6836

    July 15, 2024 at 8:12 am

    So this is how they make Skynet, by training it on real world phenomena.

    • @myparceltape1169

      July 16, 2024 at 11:08 am

      The mice were very annoyed when the computer called earth was dismantled 2 weeks too early

  13. @radiumminis

    July 15, 2024 at 8:18 am

    Ai like this smells like snake oil. Why is Ted talk giving this a platform?

  14. @kaivalyashah

    July 15, 2024 at 9:02 am

    FINALLY somebody who is talking beyond dumb language models!!! I’m long on this company and DeepMind

  15. @asadbanglesandcosmetics

    July 15, 2024 at 9:20 am

    May the almighty fullfill all your dreams. May he help you be brave and face problems easily.

  16. @Incompletegaming32

    July 15, 2024 at 9:45 am

    ‌

  17. @dougg1075

    July 15, 2024 at 10:20 am

    If we can make it past the filter we are up against.

  18. @AdvantestInc

    July 15, 2024 at 10:46 am

    What a brilliant presentation by Anima Anandkumar! Her exploration of AI’s role in transforming science and engineering, especially in applications like medical catheter design, is truly enlightening.

    • @Prof_Anima_Anandkumar

      July 17, 2024 at 4:33 pm

      Thank you!

  19. @WeylandLabs

    July 15, 2024 at 10:55 am

    I feel it is frustrating for people who don’t have the resources already thought of everything they talk about years ago. – Am I wrong ?

  20. @zeroonetime

    July 15, 2024 at 11:08 am

    India consequently discovered the 0, after Abraham discovered the 01 Creation, and called it Yesh meAiin something out of nothing and called YH~WH

  21. @phreeesubz

    July 15, 2024 at 12:55 pm

    *I swear i was waiting for her face to open up and her admit that she was a robot*

    • @romchikyoung6863

      July 15, 2024 at 9:44 pm

      Why?

  22. @youteacher78

    July 15, 2024 at 5:24 pm

    Just stop trying to put AI in everything. It doesn’t solve any real problems but it creates new ethical dilemmas. The hype cycle will die down and people will wonder what all the fuzz was about, because in the end, a LLM is just big bag of (stolen) data that you shake really hard and something falls out. It has more in common with gambling than with intelligence. ????

  23. @brianwaltenbaugh

    July 15, 2024 at 8:02 pm

    Why are we training AI to broaden the overstimulated middle/upper class when broke m-f’ers ain’t even got hope. ????

  24. @myparceltape1169

    July 16, 2024 at 11:03 am

    To find the Answer will require a huge computer.
    I will begin designing it. It’s name will be EARTH, said Deep Thought.

    Sorry, I had to write that. Anima made such a good case for it. ????

  25. @zahidhussainsahito2462

    July 17, 2024 at 8:41 am

    Really amazing advancements

Leave a Reply

Your email address will not be published. Required fields are marked *

CNET

iPhone Duo Livestream: How Many Folds Can Apple’s Phone Take? Day 13

Apple’s iPhone Duo is out and we want to see how durable it is. So we built a robot to fold it 20,000+ times a day to test its limits. How long do you think it will last? 100,000 folds? 200,000 folds? 500,000 folds? Tune in and watch the testing live. We’ll have CNET’s editors…

Published

on

Apple’s iPhone Duo is out and we want to see how durable it is. So we built a robot to fold it 20,000+ times a day to test its limits. How long do you think it will last? 100,000 folds? 200,000 folds? 500,000 folds? Tune in and watch the testing live. We’ll have CNET’s editors online to answer any of your questions.

Join us weekdays from 10:00 AM – 2:00 PM PT / 1:00 PM – 5:00 PM ET

Read more about the iPhone Duo Folding Test on CNET.com
We Answer 16 Questions About Folding the iPhone Duo Hundreds of Thousands of Times

The iPhone Duo Is Now in Our Custom-Made Foldable Phone Test Machine. Will It Beat the Galaxy Fold 8?

iPhone Duo Review: Late to the Foldable Party, but It Steals the Show

Check out the 3D Printer Matthew Murray used to print the sled that iPhone Duo is sitting on:
Looking for the Most Accurate 3D Printer? Our Exclusive Testing Reveals One Clear Winner

Video Credits:
Host: Matthew Murray
Producer: Stephen Beacham
Thumbnail Designer: Zooey Liao

Add CNET as a trusted news source
Never miss a deal again! See CNET’s browser extension 👉
Check out CNET’s Amazon Storefront:
Subscribe to CNET on YouTube:
Follow us on TikTok:
Follow us on Instagram:
Follow us on Bluesky:
Like us on Facebook:
CNET’s AI Atlas:
Follow us on X:
Visit CNET.com:

#apple #iphone #iphonefold #iphoneduo #applefoldable #foldablephone #foldable #folding

Continue Reading

Science & Technology

Why the data center you dread might never even be built | Equity Podcast

The data center boom is running into serious real-world obstacles, from widespread local opposition and permitting delays to a scramble for electricity. Before those challenges even come into play, though, there’s a more basic problem: A lot of the projects being announced may never get built in the first place. With billions of dollars in…

Published

on

The data center boom is running into serious real-world obstacles, from widespread local opposition and permitting delays to a scramble for electricity. Before those challenges even come into play, though, there’s a more basic problem: A lot of the projects being announced may never get built in the first place. With billions of dollars in AI infrastructure investment at stake, separating realistic projects from speculative ones is becoming increasingly important.

On this episode of TechCrunch’s Equity podcast, hosts Rebecca Bellan and Tim De Chant are joined by Brent Nelson, senior managing director of markets and strategy at Ascend Analytics, to separate the data center projects that are likely to get built from the ones that may never get off the ground.

Subscribe to Equity on YouTube, Apple Podcasts, Overcast, Spotify and all the casts. You also can follow Equity on X and Threads, at @EquityPod.

Chapters:
00:00 Intro
2:20 Will the midterms and data center backlash slow the buildout?
5:33 Announced vs. actually built: the speculation problem
10:01 Who’s lending, and why tenants matter most
14:59 Powering data centers: why everyone defaults to gas
18:44 The case for curtailable interconnections plus batteries
20:09 Why the grid can’t just move power around
24:46 Can data centers pay their way out of opposition?
27:22 What founders and investors shouldn’t assume
29:31 Permitting reform and the lame-duck window
31:17 Outro

Continue Reading

Bloomberg Technology

Odd Lots: The AI Trade Shrugs Off a Blue Wave and a Tech Scandal

It’s been a big month so far for American politics and the tech industry. Democrats deliver a sweeping win across the country, calling continued AI development into question. Nvidia launches an investigation into CEO Jensen Huang’s affair with an alleged Chinese spy. There are plenty of things to make investors jittery, yet we’re seeing a…

Published

on

It’s been a big month so far for American politics and the tech industry. Democrats deliver a sweeping win across the country, calling continued AI development into question. Nvidia launches an investigation into CEO Jensen Huang’s affair with an alleged Chinese spy. There are plenty of things to make investors jittery, yet we’re seeing a tech rally as markets await the upcoming Anthropic IPO.

Bloomberg’s Michael Ball joins Joe Weisenthal and Tracy Alloway on the Odd Lots podcast to unpack why the tech trade is showing resilience despite macro jitters.
——–
Like this video? Subscribe to Bloomberg Tech on YouTube:

 
Watch the latest full episodes of “Bloomberg Tech” with Ed Ludlow here:

 
Get the latest in tech from Silicon Valley and around the world here:

Connect with us on…
X:
Facebook:
Instagram:
 
Follow Ed Ludlow on X here:
 
Listen to the daily Bloomberg Tech podcast here:

 
More from Bloomberg Business
Connect with us on…
X:
Facebook:
Instagram:
LinkedIn:
TikTok:

Continue Reading

Trending