Chat GPT

[bluejon]

Interesting you should say that. So I tried it:

View attachment 72416

Seems to me it DOES understand cause and effect. Certainly in this exactly it seems to, anyway.

Do you have any other tests?

no it doesn't understand it just knows how to respond with the answer above. it doesn't understand what pressure is, and how it is related to the barometer, and that causality only goes one way. it's just recognised other pieces of work saying the same thing so responds in the same way.

 

[mindmyp's_n_q's]

Are they moral? Could you ask them the best way to commit a crime?

 

[york away to this!]

Only if it has untold data about various crimes, what the results and consequences were of those crimes, and some way of being informed about the parameters of "the best way" to commit them.

 

[york away to this!]

I suspect it could suggest a number of ways that scousers should not be held responsible for said crimes...

 

[Chippy_boy]

no it doesn't understand it just knows how to respond with the answer above. it doesn't understand what pressure is, and how it is related to the barometer, and that causality only goes one way. it's just recognised other pieces of work saying the same thing so responds in the same way.

Well it disproved your point for a start. Why is humility such a rare trait on this forum I wonder? Where's the "ok that's interesting Chippy, I didn't realise it could do that now"? Is this an insult to your manhood or something?

 

[bluejon]

Well it disproved your point for a start. Why is humility such a rare trait on this forum I wonder? Where's the "ok that's interesting Chippy, I didn't realise it could do that now"? Is this an insult to your manhood or something?

It hasn't disproved my point at all, you don't understand what you are talking about and are refusing to listen to someone who does this for a living. I'm trying to explain that it is just capturing what it has found to be similar to the query you put into it and has synthesized an answer from that material. ChatGPT is not capable of original thought or understanding and it cannot understand causality. When a parrot 'speaks' it doesn't understand what it means, it just knows how to make the noise.

 

[Chippy_boy]

It hasn't disproved my point at all, you don't understand what you are talking about and are refusing to listen to someone who does this for a living. I'm trying to explain that it is just capturing what it has found to be similar to the query you put into it and has synthesized an answer from that material. ChatGPT is not capable of original thought or understanding and it cannot understand causality. When a parrot 'speaks' it doesn't understand what it means, it just knows how to make the noise.

No. You're just being a wanker who has zero humility. And regards your profession, you're obviously shit at it.

You specifically gave an example of how to prove that AI doesn't understand cause and effect. I used YOUR example as a test. Not mine, not someone else's, YOURS. And it immediately disproved or at least contradicted your point.

What you SHOULD have said was wow that's interesting but no, you just decided to act like a dick. Consistent with pretty much every post you've made on this thread. Maybe when you reflect on this you'll realise your attitude stinks, but I doubt it.

At the very least you should have said, "ok perhaps my test wasn't the best"

And btw, you are just plain wrong about much of this technology. You're miles out of date. I would ask you to explain to me how a language model can solve logic tests it has never heard before, if all it is doing is essentially predictive text. But I won't ask you that because (a) you have no fucking clue and (b) I don't give a shit what you think. You prat.

Try asking a parrot to solve a complex logic problem FFS.

Redirect Notice

 

[Franny 1234]

I think its brilliant.
The creative amongst us will coin it in if we can find a niche.

I asked it to write me a short essay about Manchester City since 2008.

This is the return in less than 15 seconds

Manchester City Football Club has undergone a remarkable transformation since 2008. Prior to that year, the club had a reputation for underachievement and inconsistency in the English football leagues. However, in 2008, the club was taken over by Sheikh Mansour, a member of the royal family of Abu Dhabi, and this marked the beginning of a new era for Manchester City.

Under the ownership of Sheikh Mansour, Manchester City has become one of the most successful football clubs in the world. The club has invested heavily in players, coaches, and infrastructure, and this has paid off with a string of domestic and international trophies. Since 2008, Manchester City has won five Premier League titles, six League Cups, and one FA Cup. In addition, the club has reached the final of the UEFA Champions League for the first time in its history in 2021.

One of the key factors behind Manchester City's success has been the appointment of top-class coaches. In 2008, the club hired Mark Hughes as its manager, but after a disappointing start to the season, he was replaced by Roberto Mancini. Mancini led the club to its first major trophy in 35 years, the 2011 FA Cup, and followed it up with the club's first Premier League title in 2012. After Mancini, the club appointed Manuel Pellegrini, who won the Premier League and two League Cups during his three-year tenure. Since 2016, the club has been managed by Pep Guardiola, widely regarded as one of the best coaches in the world, and he has led the team to unprecedented success.

Manchester City's success has also been built on a strong squad of players. The club has invested heavily in top-class talent, such as Sergio Aguero, David Silva, Vincent Kompany, and Kevin De Bruyne, and has also developed young players through its academy. The team plays an attacking, possession-based style of football that has won plaudits from fans and experts alike.

Off the pitch, Manchester City has also made significant investments in infrastructure. In 2011, the club moved to a new state-of-the-art training facility, the City Football Academy, which has helped to develop young players and improve the fitness and performance of the senior squad. The club has also invested in its stadium, the Etihad Stadium, and has expanded its capacity to over 55,000 seats.

In conclusion, Manchester City has undergone a remarkable transformation since 2008, and has become one of the



it cut short at this point but I am suitably impressed with it.

How many fa cup wins ?

 

[bluejon]

No. You're just being a wanker who has zero humility. And regards your profession, you're obviously shit at it.

You specifically gave an example of how to prove that AI doesn't understand cause and effect. I used YOUR example as a test. Not mine, not someone else's, YOURS. And it immediately disproved or at least contradicted your point.

What you SHOULD have said was wow that's interesting but no, you just decided to act like a dick. Consistent with pretty much every post you've made on this thread. Maybe when you reflect on this you'll realise your attitude stinks, but I doubt it.

At the very least you should have said, "ok perhaps my test wasn't the best"

And btw, you are just plain wrong about much of this technology. You're miles out of date. I would ask you to explain to me how a language model can solve logic tests it has never heard before, if all it is doing is essentially predictive text. But I won't ask you that because (a) you have no fucking clue and (b) I don't give a shit what you think. You prat.

Try asking a parrot to solve a complex logic problem FFS.

Redirect Notice

The example you're so chuffed you've disproved was me paraphrasing a section from the intro of Judea Pearl's book, The Book of Why. Give it a google. And if you reckon you're still onto something, seriously, you need to get in touch with him. You might have found your new calling in life.

 

[bakerdave76]

Is any one Following this ?
I gave GPT-4 a budget of $100 and told it to make as much money as possible.

I'm acting as its human liaison, buying anything it says to.

Do you think it will make smart investments and build an online business?

Follow along

 

[Chippy_boy]

The example you're so chuffed you've disproved was me paraphrasing a section from the intro of Judea Pearl's book, The Book of Why. Give it a google. And if you reckon you're still onto something, seriously, you need to get in touch with him. You might have found your new calling in life.

Quote from you: "The main difference is causality. We understand causality naturally, AI in its current form does not. An example I use with my students is a pressure barometer. When the atmospheric pressure changes the needle on the barometer moves, but if we move the needle on a barometer the air pressure does not change. AI doesn't understand this, it just learns the two are linked."

So, as this clearly does not work as a test, as I just showed you, why did you not say "Hmmmmm, OK perhaps I should start using a different analogy"? Why just the strident attitude?

And what about the link I just posted telling you about how some AI's can now understand cause and effect? Why didn't you say "Wow, I didn't realise that (obviously)". Bizarre to me that someone should be so confrontational from the off, and continue in the same vein.

 

[Gorton_Tubster]

Quote from you: "The main difference is causality. We understand causality naturally, AI in its current form does not. An example I use with my students is a pressure barometer. When the atmospheric pressure changes the needle on the barometer moves, but if we move the needle on a barometer the air pressure does not change. AI doesn't understand this, it just learns the two are linked."

So, as this clearly does not work as a test, as I just showed you, why did you not say "Hmmmmm, OK perhaps I should start using a different analogy"? Why just the strident attitude?

And what about the link I just posted telling you about how some AI's can now understand cause and effect? Why didn't you say "Wow, I didn't realise that (obviously)". Bizarre to me that someone should be so confrontational from the off, and continue in the same vein.

This guy only had a different point view from you (And he works in this field).... Since then you've called him a wanker, has no humility, the whole forum has no humility, and that he's shit at his profession!

Other than that I hardly read a word of your argument, all I saw was how shit you were :(

 

[jitsubluedan]

How many fa cup wins ?

Spotted that myself - not as reliable at picking up all the facts as we are led to believe, it seems!

 

[west didsblue]

I think its brilliant.
The creative amongst us will coin it in if we can find a niche.

I asked it to write me a short essay about Manchester City since 2008.

This is the return in less than 15 seconds

Manchester City Football Club has undergone a remarkable transformation since 2008. Prior to that year, the club had a reputation for underachievement and inconsistency in the English football leagues. However, in 2008, the club was taken over by Sheikh Mansour, a member of the royal family of Abu Dhabi, and this marked the beginning of a new era for Manchester City.

Under the ownership of Sheikh Mansour, Manchester City has become one of the most successful football clubs in the world. The club has invested heavily in players, coaches, and infrastructure, and this has paid off with a string of domestic and international trophies. Since 2008, Manchester City has won five Premier League titles, six League Cups, and one FA Cup. In addition, the club has reached the final of the UEFA Champions League for the first time in its history in 2021.

One of the key factors behind Manchester City's success has been the appointment of top-class coaches. In 2008, the club hired Mark Hughes as its manager, but after a disappointing start to the season, he was replaced by Roberto Mancini. Mancini led the club to its first major trophy in 35 years, the 2011 FA Cup, and followed it up with the club's first Premier League title in 2012. After Mancini, the club appointed Manuel Pellegrini, who won the Premier League and two League Cups during his three-year tenure. Since 2016, the club has been managed by Pep Guardiola, widely regarded as one of the best coaches in the world, and he has led the team to unprecedented success.

Manchester City's success has also been built on a strong squad of players. The club has invested heavily in top-class talent, such as Sergio Aguero, David Silva, Vincent Kompany, and Kevin De Bruyne, and has also developed young players through its academy. The team plays an attacking, possession-based style of football that has won plaudits from fans and experts alike.

Off the pitch, Manchester City has also made significant investments in infrastructure. In 2011, the club moved to a new state-of-the-art training facility, the City Football Academy, which has helped to develop young players and improve the fitness and performance of the senior squad. The club has also invested in its stadium, the Etihad Stadium, and has expanded its capacity to over 55,000 seats.

In conclusion, Manchester City has undergone a remarkable transformation since 2008, and has become one of the



it cut short at this point but I am suitably impressed with it.

It got the date of the move to the CFA wrong.
The thick fucker ;-)

 

[Franny 1234]

Spotted that myself - not as reliable at picking up all the facts as we are led to believe, it seems!

That or its programmed to down grade us. LOL

 

[Chippy_boy]

This guy only had a different point view from you (And he works in this field).... Since then you've called him a wanker, has no humility, the whole forum has no humility, and that he's shit at his profession!

Other than that I hardly read a word of your argument, all I saw was how shit you were :(

OK my post was probably a bit OTT but he was really pissing me off with the lack of humility and condescending attitude, which is poor behaviour in my book.

Perhaps you should look at the preceding posts before jumping to conclusions. I started out with saying I respect his opinion of course, and his very first reply was to dismiss everything I had posted without even reading it or watching fully any of the links.

I don't mind (so much) being lectured at when someone knows what they are talking about, but when someone claims to have expertise, and then makes statements which are questionable at best, and instead of saying "OK that's interesting", just comes back with condescending shite (again) then what you do you expect?

I'd like to think if I post something which may not be right, instead of just dismissing someone else's post I might actually say "Oh, I didn't realise that", or "hmmmm, that's interesting". Not "You don't know what you're talking about".

 

[Chippy_boy]

BTW (and I'll post this separately because it really is a completely separate point), it's absolutely inconceivable to me that these latest large language models do not "understand" anything. Sure they may not have understanding in the sense that we humans do, but at some level they do figure out what is going on. It cannot be solely pattern recognition and word matching/prediction.

I think a good demonstration is to come up with a puzzle which you have never seen or heard before and to ask e.g. GPT4 about it. So there is no way it can simply look into its database and find the answer. It has to figure it out for itself. Here, I'll try it out:

"Paul is a decorator and can paint a typical lounge in around 4 hours. Or he can paint a kitchen in 2 hours. It's 10:00 in the morning and Paul was looking forward to a pizza for lunch. Should he paint the lounge first or the kitchen first"?

I've just made this up. GPT4 has never seen this question before, so how does it do?

GPT4: "
As Paul is a decorator, he might want to choose the task that fits his schedule better before his lunch break. Since it's currently 10:00 am, let's consider how long each room will take to paint.
If Paul starts painting the lounge, which takes around 4 hours, he will finish around 2:00 pm. If he starts with the kitchen, which takes around 2 hours, he will finish around 12:00 pm.
If Paul is looking forward to having lunch, it would be better for him to paint the kitchen first. This way, he can finish the kitchen by 12:00 pm and enjoy his pizza lunch break. After lunch, he can then continue with painting the lounge."

Now if anyone tells me that GPT4 does not at some level "understand" and is simply predicting likely next words, I find that extremely hard to believe.

 

[I'm With Stupid]

Is any one Following this ?
I gave GPT-4 a budget of $100 and told it to make as much money as possible.

I'm acting as its human liaison, buying anything it says to.

Do you think it will make smart investments and build an online business?

Follow along

Fun idea although it looks like so far it’s just “put it on Twitter and get loads of sponsorship deals” which isn’t really in the spirit of it. I might try it myself with a tenner.

 

[SkyBlueFlux]

Fun idea although it looks like so far it’s just “put it on Twitter and get loads of sponsorship deals” which isn’t really in the spirit of it. I might try it myself with a tenner.

Agree he came up with a really good idea and completely sold it out immediately for social clout.

If you want to do this properly, you need to give it enough resources that it can actually do something (maybe like £1,000) and then don't tell a soul about it until it's already at the very least breaking even.

 

[Bigga]

BTW (and I'll post this separately because it really is a completely separate point), it's absolutely inconceivable to me that these latest large language models do not "understand" anything. Sure they may not have understanding in the sense that we humans do, but at some level they do figure out what is going on. It cannot be solely pattern recognition and word matching/prediction.

I think a good demonstration is to come up with a puzzle which you have never seen or heard before and to ask e.g. GPT4 about it. So there is no way it can simply look into its database and find the answer. It has to figure it out for itself. Here, I'll try it out:

"Paul is a decorator and can paint a typical lounge in around 4 hours. Or he can paint a kitchen in 2 hours. It's 10:00 in the morning and Paul was looking forward to a pizza for lunch. Should he paint the lounge first or the kitchen first"?

I've just made this up. GPT4 has never seen this question before, so how does it do?

GPT4: "
As Paul is a decorator, he might want to choose the task that fits his schedule better before his lunch break. Since it's currently 10:00 am, let's consider how long each room will take to paint.
If Paul starts painting the lounge, which takes around 4 hours, he will finish around 2:00 pm. If he starts with the kitchen, which takes around 2 hours, he will finish around 12:00 pm.
If Paul is looking forward to having lunch, it would be better for him to paint the kitchen first. This way, he can finish the kitchen by 12:00 pm and enjoy his pizza lunch break. After lunch, he can then continue with painting the lounge."

Now if anyone tells me that GPT4 does not at some level "understand" and is simply predicting likely next words, I find that extremely hard to believe.

GPT-3 talks less...