ChatGPT Results: Testing Special Relativity

Click For Summary

Discussion Overview

The discussion revolves around the implications of AI, specifically ChatGPT, on job security in various fields, with a focus on the teaching profession and the accuracy of AI in solving problems related to Special Relativity. Participants explore the capabilities and limitations of AI in understanding and processing complex concepts.

Discussion Character

  • Debate/contested
  • Exploratory
  • Technical explanation

Main Points Raised

  • One participant shares a problem involving Special Relativity and critiques ChatGPT's response, suggesting it lacks understanding despite its authoritative tone.
  • Another participant expresses skepticism about the current capabilities of AI, indicating that while tools like the Wolfram Alpha plugin may enhance user experience, they do not necessarily improve problem-solving abilities.
  • Some participants argue that AI will lead to job reductions, particularly for roles that can be automated, while others suggest that workers will transition to different roles rather than face unemployment.
  • A participant emphasizes the irreplaceable nature of human teachers in the humanities, arguing that the interaction and judgment required in teaching cannot be replicated by AI.
  • Concerns are raised about the limitations of self-learning AI and the current state of technology in replacing human roles in creative fields.

Areas of Agreement / Disagreement

Participants generally agree that AI will impact job availability, but there is disagreement on the extent and nature of this impact, particularly regarding the irreplaceability of certain professions like teaching. The discussion remains unresolved regarding the capabilities of AI in solving complex problems accurately.

Contextual Notes

Participants express uncertainty about the future capabilities of AI and its implications for job markets, highlighting the dependence on definitions of job roles and the evolving nature of work due to automation.

epovo
Messages
114
Reaction score
21
Hi guys,
I have heard a lot of claims about ChatGPT and how it is going to put a lot of people out of work. I have a friend who is a teacher of humanities and he is fearful of losing his job in the long term. People seem to be in awe at the thing and think it understand what is saying. So I decided to give it a go with a simple problem in Special Relativity and see what it comes out. It's funny because of the contrast between the authoritative tone it uses and the load of crap it produces. Judge for yourselves!

Q: An observer A sees a second observer B move along the x-axis at 90% of the speed of light. B sees an object O move along the y-axis at 80% of the speed of light. What is the speed and direction of O as measured by A?

A: To find the speed and direction of object O as measured by observer A, we can use the relativistic velocity addition formula:

v' = (v + u) / (1 + vu/c^2)

where v is the velocity of observer B relative to observer A along the x-axis (0.9c), u is the velocity of object O relative to observer B along the y-axis (0.8c), and c is the speed of light.

Substituting these values into the formula, we get:

v' = (0.9c + 0.8c) / (1 + 0.9c * 0.8c/c^2) = 1.64c / 1.72 = 0.9535c

Therefore, object O is moving at a speed of 0.9535 times the speed of light as measured by observer A. To determine the direction of O's motion as measured by observer A, we can use the Lorentz transformation equations as before:

x' = γ(x - vt)
y' = y
z' = γ(z - vt)
t' = γ(t - vx/c^2)

Since observer B sees object O moving only along the y-axis, we can set x = z = 0 in the transformation equations. Solving for y' and t', we get:

y' = y
t' = γ(t - vx/c^2)

Therefore, object O is moving perpendicular to the x-axis in observer A's frame of reference, i.e., along the y-axis. The negative sign in the previous answer indicates that O is moving in the opposite direction to B's motion along the x-axis, so we can conclude that object O is moving in the negative y-direction as measured by observer A.
 
  • Like
Likes   Reactions: Greg Bernhardt
Physics news on Phys.org
It's definitely not "there" yet for many cases, but imagine 5 years from now. Note that Wolfram Alpha just released an early access plugin for ChatGPT. That will immediately level up the calculation power.
 
  • Like
  • Wow
Likes   Reactions: topsquark, russ_watters and Wrichik Basu
I think this demonstrates a complete lack of understanding, which is not surprising being, as it is, a language model. If I am not mistaken the Wolfram Alpha plugin intends to make life easier for the user when expressing the problem that they want Wolfram Alpha to solve, not to enhance the problem-solving capabilities of the tool.
 
  • Like
Likes   Reactions: topsquark, russ_watters and Greg Bernhardt
The threat at least near term of AI taking jobs is not that AI as an independent agent will do it but rather that those that can use AI effectively will. In any event, AI will reduce the number of jobs currently available.
 
  • Like
Likes   Reactions: Filip Larsen, russ_watters and epovo
I agree that the threat is real for certain jobs
 
gleem said:
AI will reduce the number of jobs currently available.
I had to read that a few times to be sure I agreed, and it depends on what you mean by "currently available". Yes, it will reduce the number of workers needed to accomplish certain tasks today, as automation always has. So when looking from today there will be fewer "current" workers needed. But of course they won't be unemployed, they will just get different jobs or be more productive, the same effect automation has always had.
epovo said:
Hi guys,
I have heard a lot of claims about ChatGPT and how it is going to put a lot of people out of work. I have a friend who is a teacher of humanities and he is fearful of losing his job in the long term.
I don't see what one thing has to do with the other. The job of a humanities teacher is to teach other humans art, literature, etc. and judge their understanding (while the students interact with them and other human students). That job inherently has to be done by a human teacher, in a classroom, with a group of human students.

You can't replace that with a chat-bot. At the very least you need a true human-replacement sentient AI.

It's similar to one of the the reasons why movies still use human actors. It matters to audiences that you're watching Sigourney Weaver and not just a cartoon that looks a little like her (and yes, the simulations arent quite ready yet either).

That's the flaw in self-learning we constantly point out in the academic forums, and so far nothing I've seen implies a change on the horizon.
 
Last edited:

Similar threads

  • · Replies 190 ·
7
Replies
190
Views
16K
  • · Replies 6 ·
Replies
6
Views
989
  • · Replies 25 ·
Replies
25
Views
1K
  • · Replies 34 ·
2
Replies
34
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 21 ·
Replies
21
Views
2K
  • · Replies 26 ·
Replies
26
Views
4K
  • · Replies 33 ·
2
Replies
33
Views
2K
  • · Replies 20 ·
Replies
20
Views
2K
  • · Replies 13 ·
Replies
13
Views
2K