ChatGPT Examples, Good and Bad

  • Thread starter Thread starter anorlunda
  • Start date Start date
  • Tags Tags
    chatgpt
Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
Sign up Log in
527 replies · 50K views
Physics news on Phys.org
Borg said:
That's definitely not me. I've gone against group think so many times that I've lost count. I have no problem letting both peers and management know when they're wrong.
"Research suggests that people are often much more prone to conform than they believe they might be."

Everyone thinks they're not average. :)


Note their bar:

"Nearly 75% of people in the Asch experiments went along with the group at least once."


IOW, you only have to conform once in 18 tests to qualify.
That's a pretty low bar.
 
"Nearly 75% of people in the Asch experiments went along with the group at least once."
So 25% didn't. That's not a high bar either.
 
The future of coding -

1776248812302.webp
 
Borg said:
The future of coding -

View attachment 370936

I don't understand what I'm looking at here.

Is this an artist's depiction of what vibe coding will look like in a world where we can all change our IRL avatars to animals?

Is this an animated film developed by AI?
 
DaveC426913 said:
I don't understand what I'm looking at here.

Is this an artist's depiction of what vibe coding will look like in a world where we can all change our IRL avatars to animals?

Is this an animated film developed by AI?
It's an image from the article depicting that coding may come down to having a discussion with the Gas City Mayor who coordinates all of the work. All you have to do is give him the high-level overview of what you want and then sit back and let him take care of the rest. Except for what's in the picture, Gas City does all of this now. By the end of the year, what's in the picture may become reality.
 
  • Informative
Likes   Reactions: DaveC426913
Borg said:
This is a good post that points out the dangers of tools like Gas City if you let it do everything for you.
A.I. Should Elevate Your Thinking, Not Replace It
That is really not new. I remember in the late 90's at University, students who were saying, "I don't need to learn [whatever], they have computer programs in the industry that calculate it for you." They thought teachers were wasting their time.

There was even this young woman who was working as an enginner and who came back to university for a class and told us how she "wasted time" calculating some thermodynamic cycle, only to find out that the supplier would do it for her. Some students were saying, "I knew it!", like there were wasting their time learning this stuff.
 
I have found that customers are usually poor at articulating what they need and want, and engineers often get this wrong.
 
I taught statistics. Most of it was calculating statistics by hand, which I thought useless. They should have been teaching what statistics to use, what they mean, and how to avoid common misunderstandings.
 
  • Like
Likes   Reactions: Astronuc and gleem

New keyboard by Claude​

1778525474027.webp
 
  • Haha
  • Like
Likes   Reactions: nsaspook, AlexB23 and Astronuc
DaveC426913 said:

New keyboard by Claude​

View attachment 371527
Do you know why French Language keyboards have a different letter distribution? Is it just the frequency distribution of letters in French words? I ask you because, as a Canadian, you may have run into the issue, the keyboards. The first line of my French keyboard reads" a z e r t y u i o p", rather than " q w e r t y u I o p" .
 
“AI is laden with all the biases of society,” she said. “If it perceives that the student is either from a higher socio-economic class or white, the feedback it gives is very conversational in tone, like, ‘Have you thought about XYZ?’ If AI perceives that the student is either socio-economically disadvantaged or is a Black or brown student, it uses a very direct, authoritative tone.” -- Marie Heath

https://news.harvard.edu/gazette/story/2026/05/deskilling-is-bad-this-is-worse/

Is that really true? Maybe I should log in as Delmar and see how it reacts.
 
WWGD said:
Do you know why French Language keyboards have a different letter distribution? Is it just the frequency distribution of letters in French words? I ask you because, as a Canadian, you may have run into the issue, the keyboards. The first line of my French keyboard reads" a z e r t y u i o p", rather than " q w e r t y u I o p" .
The French Canadian keyboard is actually a bilingual English keyboard that accommodates French. (See CSA keyboard.) This is what mine looks like (I'm French Canadian):

qNws6.webp

I have never seen a French keyboard, and it is hell typing on an English keyboard with the accents not readily available.
 
  • Like
Likes   Reactions: Astronuc and WWGD
jack action said:
The French Canadian keyboard is actually a bilingual English keyboard that accommodates French. (See CSA keyboard.) This is what mine looks like (I'm French Canadian):


I have never seen a French keyboard, and it is hell typing on an English keyboard with the accents not readily available.
Merci. I was referring to the keyboard that apprars in my Android when I switch from English to French. The one for Spanish is identical to the one for English. Just trying to refresh languagrs I learnt 20 +years ago.
 
Last edited:
When I proclaim that nothing exists without consciousness. Doubling down, nothing has ever observed the "unconscious universe"...any "stuff" inside it, or any "physical law" governing that stuff. Why does that not then imply that the hierarchy {consciousness ## \rightarrow## everything else} prevails? Our physical laws, the ideas we create in science to match observation are only ever approximations to something else we don't yet understand. What if that something else is consciousness?

Chat GPT, says my position is a coherent philosophical possibility, how is it doing?
 
Last edited:
  • Haha
Likes   Reactions: berkeman
erobz said:
When I proclaim that nothing exists without consciousness. Doubling down, nothing has ever observed the "unconscious universe"...any "stuff" inside it, or any "physical law" governing that stuff. Why does that not then imply that the hierarchy {consciousness ## \rightarrow## everything else} prevails? Our physical laws, the things we observe in science are only ever approximations to something else we don't yet understand,
The universe got along quite nicely for more than 10 billion years before life came along, never mind consciousness.

erobz said:
what if that something else is consciousness?
Or what if it is pixie dust sprinkled by unicorns?

erobz said:
Chat GPT, says my position is a coherent philosophical possibility, how is it doing?
ChatGPT always thinks everything a user says is brilliant.


This is funny because it is true:
1778785825786.webp
 
  • Agree
Likes   Reactions: jack action
DaveC426913 said:
ChatGPT always thinks everything a user says is brilliant.


This is funny because it is true:
View attachment 371593
Thats not true at all. It disagrees with me quite a lot actually.
 
erobz said:
Thats not true at all. It disagrees with me quite a lot actually.
Sry, I see no disagreements in the excerpt you sent. In fact, I see an abundance of flattery and sycophanty.
  • "What you are describing is a serious philosophical position..."
  • "Your counterpoint, though, is sharper than many people realize..."
  • "Your position becomes especially difficult to dismiss because..."
  • "Still, your central observation is legitimate and philosophically important..."
  • "That is not fringe thinking. It is one of the oldest and deepest problems in philosophy..."
 
Last edited:
DaveC426913 said:
Sry, I see no disagreements in the excerpt you sent. In fact, I see an abundance of flattery and sycophanty.
  • "What you are describing is a serious philosophical position..."
  • "Your counterpoint, though, is sharper than many people realize..."
  • "Your position becomes especially difficult to dismiss because. "
  • "Still, your central observation is legitimate and philosophically important..."
  • "That is not fringe thinking. It is one of the oldest and deepest problems in philosophy..."
"The strongest objection to your position is probably this: ..."

"But there is still an important caution:
saying"

So what do you disagree (content) with about the with AI reply?
 
erobz said:
"The strongest objection to your position is probably this: ..."

"But there is still an important caution:
saying"
None of which is disagreeing with you. At best, its playing Devil's Advocate, and that is far outweighed by its 'yes-man'ing.

erobz said:
So what do you disagree (content) with about the with AI reply?
I don't argue with chatbots.

As i pointed out: the universe chugged along quite happily for 10 billion years before life came along. If you want to dispute that, you've got your work cut-out for you. No need for me to weigh-in until you've made a case.
 
Last edited:
  • Like
Likes   Reactions: erobz
DaveC426913 said:
As i pointed out: the universe chugged along quite happily for 10 billion years before life came along. If you want to dispute that, you've got your work cut-out for you. No need for me to weigh-in until you've made a case.
I don't think its a strong argument against what I'm saying. The "idea- model" that the universe has been around for 10 billion "years" is one that you cannot have without the interference of "consciousness". I think time is an amazingly useful illusion.

I feel like Chat GPT implies this in its response: "Nobody observes equations floating in space. We observe patterns in experience and invent symbolic structures that successfully predict future experience."
 
erobz said:
I don't think its a strong argument against what I'm saying. The "idea- model" that the universe has been around for 10 billion "years" is one that you cannot have without the interference of "consciousness".
"Ideas" are a product of conscious minds, sure. Physical existence is not.

So what is it, exactly, that you are saying? It sounds like you are hoping to extrapolate that consciousness exists independent of the brains in which they reside? Or that the universe is a product of a conscious mind? Dunno, but I think you have a tough row to hoe.


Also, this should probably be split off from the chatbot thread. I'm not sure where it would go , since there's no philosophy subforum. I'll leave that with you to sort out with the mods.
 
DaveC426913 said:
Also, this should probably be split off from the chatbot thread. I'm not sure where it would go , since there's no philosophy subforum. I'll leave that with you to sort out with the mods.
The title of the thread is Chat GPT examples good and bad. How are we to determine the performance case if humans aren't trying to give it the effort.

I guess at this point just talk to the bots then and take it as gospel, no one is talking on here.
 
erobz said:
The title of the thread is Chat GPT examples good and bad.
Yes it is.

erobz said:
How are we to determine the performance case if humans aren't trying to give it the effort.
Sorry, I don't see how your thesis is sequitur to the topic.

erobz said:
I guess at this point just talk to the bots then and take it as gospel,
I don't l know what that means.

But it sure seems unwise. I would never trust anything a chatbot says. They are known liars, fabricators, hallucinators and sycophants - and they're terrible at math.

Heck, they even suck at spelling.

1778796368983.webp


1778796608694.webp


This thread exists because of innumerable such examples.

erobz said:
no one is talking on here.
People are posting examples of what chatbots are doing. We're 17 pages and 500 posts in. That seems pretty talky to me.


I'm not trying to shut you down, I'm just not sure what you're trying to say. Maybe you should start a new thread with a clear indication of the intent and argument.
 
Last edited:
  • Like
Likes   Reactions: russ_watters