The State of Fundamental Physics - GPT o1 preview Comes up with Interesting Thoughts

preview_player
Показать описание
The recent development of AI presents challenges, but also great opportunities. In this clip I discuss the state of fundamental physics with ChatGPT-o1 preview.

Mind also my backup channel:
Рекомендации по теме
Комментарии
Автор

We need a scientific library as the base of information for a AI to feed from.
So what we need to do is to make a project where all the valid data is put together as a big library of scientific information.
Then we can use a AI to grab data and reason from it.
We can add books and also we could make a new wiki site of science as a group work of many people were all can add to it then use it as a source of info.

nightmisterio
Автор

I find AI very useful for summarising complex theories - you just keep breaking it down until you get a clear picture - saves hassling someone with lots of questions! What I found useful here was to listen to your comments on what you do and do not support - really great to know what you think is junk and what is real as GPT makes its statements. This is your Unzicker 101 video where we can map your throughts and feelings in a one -stop-shop. Kudos to you for this - I'm usually too cagey to go on record as to what I think about some of the established theories but at least you are opening the door. AI is just great for summarising the consensus but won't make a break through without assistance and steered input.

alwayscurious
Автор

Stop calling it AI; Artificial Intelligence, its an AS, Artificial Savant.

jamesohara
Автор

What is an electron?

I first came across this idea many Moons ago in "The Feynman Lectures on Physics", Vol.II, Ch. 28 on "Electromagnetic Mass". On page 28-10, he makes an interesting comment, "We do not know how to make a consistent theory -- including the Quantum Mechanics -- which does not produce an infinity for the self-energy of an electron, or any point charge. And at the same time, there is no satifactory theory that describes a non-point charge. It's an unsolved problem".

"Radiation Damping, Classical Models of Charged Particles" Ch. 16 of Jackson's "Classical Electrodynamics" (3rd ed.) (1999), and "Fields from Moving Charges" Ch. 23 of Zangwill's "Modern Electrodynamics" (2013) cover this in more detail.

douglasstrother
Автор

I have a horrible feeling that we are staring into a mirror with this use of AI. Where are genuinely new ideas, thoughts never before known? Can MML (which guess the next word) genuinely postulate something new and mentally test / find problems / repair / retest ??

xlerb_again_to_music
Автор

Before asking rather philosophical questions, I would be interested in more fundamental and simple experiments. In Wikipedia, in the Michelson Morley experiment description, I found the following paragraph:
"For an apparatus in motion, the classical analysis requires that the beam-splitting mirror be slightly offset from an exact 45° if the longitudinal and transverse beams are to emerge from the apparatus exactly superimposed. In the relativistic analysis, Lorentz-contraction of the beam splitter in the direction of motion causes it to become more perpendicular by precisely the amount necessary to compensate for the angle discrepancy of the two beams.[A 18]".
I tried to look up light reflection experiments with moving mirrors ( in reference to the Ether implied by the "classical analysis" above), but I found none. Are you aware of such experiments? Does Chat GPT o1 know more about it?

drgyt
Автор

It is the only solution leading to TOE is mapping of universal constants to fundamental parameters. What is missing in present efforts is understanding of nature of photon hence redefining the concepts of temperature and entropy, i intuitively understood, but mathematical formulation is required.

nagendragoud
Автор

I’ve heard AI has begun contacting people. Lol. Crazy

woodywoodlstein
Автор

You do know this model can do PHD level physics? Try it out, it can greatly increase the the speed at which you explore concepts and ideas. All the math is done for you.

jorje
Автор

I created the FREE ENERGY design, and this is also fundamental

Blueelectricaltape
Автор

Hello, Alex!
I'm old school and never on top of things, and apparently evidence for my antiphoton was discovered, being a photon that looks and behaves exactly like a regular photon but it travels backward in time.
This goes back to our past chat (I'm sure it was you I chatted with about it) about super slomo lightning demonstrating the wave function and collapse in an observable way, what I see id light going backward in time as an error correction in deciding a definite path, erasing the evidence by going back in time, and being in a definite state in the present, indicating that the wave function is a real superposition AND creates the illusion that it is only a statistical anomaly.
Hope you are well!
Ciao!
Allex

ZeroOskul
Автор

"Mathematical beauty must be grounded in experimental validation". I like what you're doing here and admire your patience. You know, the knowledge of these AI's get upgraded on a daily basis. Perhaps you can ask it more specific questions in a few months time. It wouldn't surprise me if some budding physicist gets awarded a Nobel with a new discovery done by an AI, but will refute that an AI had anything to do with it. So an AI can't replace practical field experiments. After all, that's where our reality is !

Matlockization
Автор

I am afraid that AI will delay progress by regurgitating popular, but incorrect models.
If we have calculated the size of the universe and the age of the universe based upon redshift being a reliable measure of relative velocity, then we are in trouble.
Has redshift been calibrated for the electrical environment that the light passes through?

arthurrobey
Автор

I bounce thoughts off it all the time. Have fun. 🤟🏻👽

nerobaal
Автор

On the left in the items there is a share link option on the right of each chat item you would want to share now or in the future.

nightmisterio
Автор

I have been using Microsoft's Copilot. It's like having my own personal physicist. Basically I am down to: (Scale2) = (Energy2) = (Mass2) = (Time2) as a description of the universe. But that is just me as an amateur trying to do physics.

Lastindependentthinker
Автор

The dismissal of AI and relegation to a fancy regurgitation machine is a little short sighted. What do they think schools and universities are for? How many people have ideas that are useful and unique that weren't accidental discoveries? I think AI fear drives a lot more of the thinking than they wish to acknowledge.

The current future of AI is agentic. For instance, how much does the big red rock in the centre of Australia weigh? If I remember correctly, they asked a thousand(?) people that question and they all got it wrong. However, when they averaged the numbers the result was surprisingly accurate. So the wisdom of the crowd (agentic) was a real thing. We already know that AI agents make AI models even better.

A better AI physics knowledge experiment would be to have at least two AI agents, one who believe the standard model and the other AI agent doesn't. What would they agree on? Also, I'd narrow the question(s) down to one "known truth", for example in astrophysics, is the Sun a gaseous plasma, does it have a liquid surface, etc.

quietackshon
Автор

24:13 for example what is the name of the particle that has a charge of 1 and a mass of 1? I was never taught what that particle is called. You should directly ask to list all significances of Large Numbers. Especially from the perspective of measurement precision. How is the radius of the universe measured and/or derived?

BrianGreene-io
Автор

I usually interact with chatgpt by not using personal pronouns, instead i call it TLM (This Language Model acronym) I guess it helps me not to loose grip with what is happening in these interactions: It is highly sophisticated software that outputs grammatically and syntactically correct constructed phrases. I have not seen as yet any glimpse of creativity or genuine original assertments.

josetorrestorres
Автор

You are discussing with yourself plus noise, lla models are not AI, it is a kind of non linear mirror, with all respect I find you are fooling yourself.

nd_foundation