Can AI discover new laws of physics? | Max Tegmark and Lex Fridman

Lex Fridman Podcast full episode: • Max Tegmark: AI and Ph...
Please support this podcast by checking out our sponsors:
- The Jordan Harbinger Show: www.jordanharbinger.com/lex/
- Four Sigmatic: foursigmatic.com/lex and use code LexPod to get up to 60% off
- BetterHelp: betterhelp.com/lex to get 10% off
- ExpressVPN: expressvpn.com/lexpod and use code LexPod to get 3 months free
GUEST BIO:
Max Tegmark is a physicist and AI researcher at MIT.
PODCAST INFO:
Podcast website: lexfridman.com/podcast
Apple Podcasts: apple.co/2lwqZIr
Spotify: spoti.fi/2nEwCF8
RSS: lexfridman.com/feed/podcast/
Full episodes playlist: • Lex Fridman Podcast
Clips playlist: • Lex Fridman Podcast Clips
CONNECT:
- Subscribe to this KZbin channel
- Twitter: / lexfridman
- LinkedIn: / lexfridman
- Facebook: / lexfridmanpage
- Instagram: / lexfridman
- Medium: / lexfridman
- Support on Patreon: / lexfridman

[@falcodarkzz]

"anyone with a computer can type pip install feynman" - got me excited there then I went to the site and its for mac and linux only, you bastard!

[@jarrodhill9149]

Install Linux

[@falcodarkzz]

@@jarrodhill9149 Looks like I have to

[@yashkulkarni4080]

Run it under WSL

[@poco9964]

Sandbox linux in windows

[@user-hh2is9kg9j]

​@@jarrodhill9149 What data do you feed it?

[@chuckaudio3191]

Max RULES!!! Love that guy! I want to see him chat with Roger Penrose!

[@manjorbas74]

“We go around with our neural networks..”

[@RikWlz]

Computer. Tea, Earl Grey, Hot.

[@aidanhammer6968]

I feel like this is one of those "if you know you know" things but I'd like to know more

[@omar2886]

A better project came out recently by making an AI not to start from our known physics but to guess its own way out to describe the physical world phenomena following its own thinking. Thisbis brilliant as our physics doesn't have to be the only neither the perfect way to describe the universe, is just the way we came out with. So here we'll be gaining a new interpretation we could have never came up with, but also an exponentially much faster way to progress in physics and mathematics by letting AI drive our reseaches

[@aidanhammer6968]

Anton chigurh(from country for old men) + Clint Eastwood tell me I'm wrong

[@billfgxheru]

Good question

[@just-in-spacetime6674]

Good answer.

[@jaclacscatpack]

Seriously it is. It would make me rethink my entire beliefs if so.

[@matthewnicholson2699]

I love this dude!!

[@baljeetbhachu4273]

Could an AI win a Nobel in physics someday?

[@AS40143]

yes

[@matthewnicholson2699]

Why couldn’t it?

[@shaikhbakhas5105]

I guess probably yes

[@joakimolovsson7310]

If it could infer the meaning behind the discovery to humans, sure! Otherwise a human would get it for understanding the meaning behind the result that the AI had found. The issue at that later stage would be that humans would not understand the full reality and could know the dumbed down version or an approximation.

[@meenapandey1473]

No🙂 Don't think that they are human they are just a computer program.

[@just-in-spacetime6674]

You should see my dog catch tennis balls. Fast ball back at ya! I'll stop now.

[@KippiExplainsStuff]

my dog can't catch ANYTHING. Not even food. She'll simply squint hard and let it hit her face. That's my one true disappointment of her. I'm like - come on, do some Dogging! She doesn't get it. That being said, she doesn't lick faces (she has no concept of what one is supposed to do when she has a nose up against her nose) either so she's a little broken.

[@just-in-spacetime6674]

@@KippiExplainsStuff I'm sorry to hear that. Sounds like a great pet regardless.

[@myxalplyx]

@@KippiExplainsStuff Maybe she's squinting like...what the heck are you doing and why. We can't make assumptions. 😅

[@KippiExplainsStuff]

@@myxalplyx like I said - she has no concept of what's going on haha. But she's cute as hell - jet black cocker spaniel. She has an attitude at times but we love her

[@adelinaquijano1083]

I think he repair my door.

[@FrumpyPumpkin]

Yes.

[@chirathpinsara7845]

👍

[@matthewnicholson2699]

Wait what?? GalilEo would catch it!!!?? Not hit him in the face?

[@Dieselman5556]

How much would we pay

[@1111122234]

I solved Riemann's Hypothesis using an AI designed by me. (Not a joke).

[@snobgoblin1226]

Publish your proofs, see if it has any merit.

[@chenwilliam5176]

Are you kidding 🤣 An GTP LLM which only excute Pre- Training & Generatin has Ability to 「Create Law of Physics」?! 🤔 Please tell me How ChatGPT do it ❤

[@oneman5753]

Easy answer to the title: yes. Any sufficiently advanced system will be able to do so. Thinking humans or organics would only be capable is just looking in the mirror and flattering yourself. In fact, I would bet money machines would ultimately be better at it depending in how.... Predisposed they are too existing "laws" the way humans are biased. Remember, most of us begin by tearing down or ridiculing things that overturn our existing understanding or beliefs. You see that trend in humans EVERYWHERE from religion to science even in trying to teach equality and acceptance when trying to correct race or class systems to allow all to flourish. Even when these are just arbitrary constructs created by other humans (usually rich or powerful folks) for shit reasons

[@whiteeye3453]

No a. I cannot discover new theory because it is i capable of creativity

[@R.Tafolla]

An AI already hosts this podcast

[@GabrielPohl]

input points ai: {\displaystyle {\mathcal {L}}={\bar {\psi }}(i\gamma ^{\mu }D_{\mu }-m)\psi -{\frac {1}{4}}F_{\mu u }F^{\mu u },}

[@whiteeye3453]

Can a. I discovered something new? Anwser no Can a. I calcue problem yes

[@pomainli]

What more artificial laws..

[@theuniques1199]

Reality has to first believe it exist as reality to infinitely exist and to first believe in reality then reality has to infinitely exist because to first believe in reality then reality has to infinitely exist and reality has to first believe it exist as reality to infinitely exist. Real eternal recurrence as first belief of real existence and first belief of real existence as real eternal recurrence, neutrality as inverse and inverse as neutrality.

[@Lesminster]

AI network that can solve problems not knowing really how ? That seems like a hunch.

[@dilipdas5777]

We don't want this type of society

[@ChrisEli7e]

I cant listen to this man talk

[@ufcprophet40]

Is it just me or Max’s speech is annoying af. He stops all of sudden and then continues. It’s unnerving

[@marjanp]

He probably has speech impediment.

[@meninblack3585]

Hes smarter then anyone you guys will ever know. A lot of intelligent people have this including elon musk he'll stop and start and think. You kids wouldn't understand, and even it is something he has So what? The guys one of the most respected in the physic community.