So called Artificial Intelligence (AI)

[save the frogs]

8 minutes ago, scottiejohn said:

It cannot code itself!

It can only use propagation of already inbuilt code or direct real time input!

Ask ChatGPT:

 

Question: write a computer program that generates the average air pollution in bangkok

 

Answer:

Here is a Python program that generates the average air pollution in Bangkok: ```python import requests from bs4 import BeautifulSoup # Get data from website url = 'http://aqicn.org/city/bangkok/' r = requests.get(url) soup = BeautifulSoup(r.content, 'html.parser') # Find the air quality data data = soup.find_all('div', {'class': 'aqivalue'}) # Extract the average air pollution value pollution_values = [] for

[scottiejohn]

1 minute ago, Jingthing said:

You sound like someone who incorrectly thinks AI is just pimped up google. 

 

As it is marketed at present that is exactly what it is!

So called AI is just a mess of code attached to a big pile of data, which it remixes and regurgitates. You can ask ChatGPT to write you a resume, and it’ll spit out something based on the resumes in its dataset (plus whatever info you share). This is useful, it automates labour, but it’s not a sign of intelligence.

[OneMoreFarang]

26 minutes ago, scottiejohn said:

I don't think you understand how programming works!

I know how programming works. You have to be creative but there is also a lot of relative simple code which must be written. That is what the AI does now. And it does it well. That means that a software developer can work a lot faster because lots of the simple code is created by the AI.

Try it. I would be surprised if you are not impressed. 

[save the frogs]

6 minutes ago, scottiejohn said:

This is useful, it automates labour, but it’s not a sign of intelligence.

that's correct.

it is just synthesizing and regurgitating.

claims that it will have its own intelligence are still merely just claims.

so you are right. 

but Sam Altmann claims GPT4 is starting to show signs of "general intelligence".

Not sure if GPT4 is available to the public or how we could test that. 

[OneMoreFarang]

3 minutes ago, scottiejohn said:

 

As it is marketed at present that is exactly what it is!

So called AI is just a mess of code attached to a big pile of data, which it remixes and regurgitates. You can ask ChatGPT to write you a resume, and it’ll spit out something based on the resumes in its dataset (plus whatever info you share). This is useful, it automates labour, but it’s not a sign of intelligence.

ChatGPT writes programming code which it didn't just copy and paste from somewhere. Try it! Just tell it in plain English what the program should do and which programming language you prefer. And then watch! 

[scottiejohn]

4 minutes ago, josephbloggs said:

Are you an old man?

Depends on the definition of "old man" and why does that matter?

I have been working continuously with computers of all shapes and sizes since May 1970. 

Admittedly that first lot was a set of somewhat large mainframe systems! 

[scottiejohn]

Just now, OneMoreFarang said:

ChatGPT writes programming code

It does not!

[Jingthing]

At least for the forseeable future there will be the need for very high skilled humans who can use these tools expertly and make decisions on usability of the output. But they will be fewer and fewer and currently people doing lower level "intellectual" work days are numbered.

[save the frogs]

12 minutes ago, scottiejohn said:

It does not!

I don't think it writes original code.

I think it is using an existing database of code.

But not sure. 

[scottiejohn]

27 minutes ago, save the frogs said:

I don't think it writes original code.

I think it is using an existing database of code.

Correct!

[BangkokReady]

4 hours ago, scottiejohn said:

Even the actual phrase "Artificial Intelligence" is an oxymoron!

I don't think anyone hears the term "Artificial Intelligence" and believes that an actual conscious, self-aware sentient being has been created artificially.  It is merely programmed to appear that way in some cases. 

 

The artificial part means it is something that has been created rather than naturally occurring and the intelligence part means it can create things and make decisions (albeit within whatever it is programmed to do).

 

The self-driving car or the chat that appears close to being human suggests some form of intelligence as it carries out tasks or responds verbally without a human being directly involved, but it is just the programming.

[Dan O]

1 hour ago, scottiejohn said:

They cannot!

 

Yes they can. You may want to do some better research on this topic because you're way off on the capability of software development in this area. 

[rose33]

2 hours ago, OneMoreFarang said:

Today I read an article that a phycologist, who normally tests humans, tested ChatGPT with a standard IQ test. Result? Guess. 155! 

I Gave ChatGPT an IQ Test. Here's What I Discovered - Scientific American

 

Not long ago I saw this video. I would never have thought that would be possible:

 

 

Now I played myself with ChatGPT. I gave it a programming job in plain language. The initial response was already very good. And then I asked ChatGPT to change this and that and some more. Almost all answer were good or very good.

The only difference to having a skilled programmer on the other side was that ChatGPT was faster, and it was not annoyed when I asked deliberately a few stupid questions.

 

Anybody who dares to ignore this should not be surprised if suddenly the world changes around him or her. How did that happen. It happens. And fast!

 

For anybody with some time I recommend this video.

 

 

 

IQ is not the same as wisdom, and IQ is not a predictor of a person's usefulness to others  High-IQ people don't necessarily lead productive lives, nor necessarily bring peace and happiness to the people & communities around them.

 

High-IQ people tend to feel lonely and misunderstood. They often spend time alone because they prefer that over human company. They can become rather bitter and resentful people. When they do engage, it is comparably easy for them to exploit and manipulate others -- not because they are evil people, but for the practical reason that their processing speed & accessible memory is greater.

 

All other things being equal, I wouldn't  place a lot of trust in someone purely because of their high IQ. I would need more information about what they are like as a person, so I could get an overall picture.

 

 

 

 

[rose33]

1 hour ago, save the frogs said:

Ask ChatGPT:

 

Question: write a computer program that generates the average air pollution in bangkok

 

Answer:

Here is a Python program that generates the average air pollution in Bangkok: ```python import requests from bs4 import BeautifulSoup # Get data from website url = 'http://aqicn.org/city/bangkok/' r = requests.get(url) soup = BeautifulSoup(r.content, 'html.parser') # Find the air quality data data = soup.find_all('div', {'class': 'aqivalue'}) # Extract the average air pollution value pollution_values = [] for

Let's think about the type of questions/requests we should be making of ChatGPT in order to assess its abilities.

 

One, a computer program (on its own) is not a petrol car, a fire in a Mekong rice field, nor a factory, so we can't expect it to generate pollution (at average rates or otherwise).

 

On the face of it, the computer program fails 100%, because you asked it to generate air pollution and it failed to fulfil your request.

 

Two, I will assume you meant to ask the computer program to report some data to you -- data that YOU want, because YOU are in the driver's seat, testing ChatGPT.

 

The purpose of deriving a mean average is to be able to compare individual data points to that mean. Alternatively, you might want to compare the mean of one data set with the mean of another (comparable) data set.

 

It's not clear here what information you wanted from ChatGPT.

 

If you are accustomed to checking a (for example) daily air pollution statistic (e.g., from a news source, or a pollution-tracking app), then simply request today's data in the same way you'd use Google.

 

If you want a computer program to calculate an average, then you need to tell it what YOU want the data set to be. (Otherwise, you are simply being a passive recipient of someone else's creation, like when reading a book, watching the news or listening to a song).

 

To give actionable instructions to Chat GPT, you first need to think about WHY you want to see this data and what ACTION you'll take as a result of seeing it.

 

You'd need to tell ChatGPT:

 

- what's the time period from which the mean average is to be derived?

- (presumably there are sensors or similar placed asking Bangkok, testing the air.) Which sensors' data should be included in the calculation? Should any sensors' data be excluded? 

- what's the desired unit measure of air pollution ChatGPT should work with?

 

If you can't be bothered to truly test  ChatGPT's abilities, and you are more interested in observation and entertainment, then make a request something like: "ChatGPT, write some code that has something to do with calculating some kind of average air pollution in Bangkok." If you'd asked that question, then the answer you received from ChatGPT would have been appropriate.

 

BUT it wouldn't be a very useful or impressive answer, because the question wouldn't have been meaningful or difficult.

 

[rose33]

4 hours ago, OneMoreFarang said:

Today I read an article that a phycologist, who normally tests humans, tested ChatGPT with a standard IQ test. Result? Guess. 155! 

I Gave ChatGPT an IQ Test. Here's What I Discovered - Scientific American

 

Not long ago I saw this video. I would never have thought that would be possible:

 

 

Now I played myself with ChatGPT. I gave it a programming job in plain language. The initial response was already very good. And then I asked ChatGPT to change this and that and some more. Almost all answer were good or very good.

The only difference to having a skilled programmer on the other side was that ChatGPT was faster, and it was not annoyed when I asked deliberately a few stupid questions.

 

Anybody who dares to ignore this should not be surprised if suddenly the world changes around him or her. How did that happen. It happens. And fast!

 

For anybody with some time I recommend this video.

 

 

Seems to me this big change is coming, whether any individual likes it or not.

 

Whether anyone personally ignores ChatGPT or enthusiastically throws  questions at it (intelligent or unintelligent) every evening, won't make a scrap of difference.

 

Some people just like to jump onto every bandwagon.

 

I know a very high IQ medical doctor, who has published extensively and completed a Fellowship, and also done a lot of charity work, who tested negative to COVID four times in one week, and still insists ????????????that he had COVID at that time. "Based on what, Sir? A HUNCH???" (The doctor turns witchdoctor!)

Just a full on bandwagon rider. Has to share the same experiences that everyone else. Some people are just like that.

 

Makes no difference in the end -- so I think people should feel emboldened to "dare to ignore" (your words) ChatGPT and all the AI buzz, if that's their preference.

 

Nature is still there, instrumental music, animals, cooking... all the good things that were always there are still there. 

 

I for one cherish the opportunity to talk to those who still have the ability to generate THEIR OWN THOUGHTS. These days most people cannot converse; they merely regurgitate from news and social media. Like demented parrots. ChatGPT will only exacerbate the problem.

 

I should pick out a few choice individuals right now and beg them to ring fence with me ;-) The Anti ChatGPT Chat Group.

 

[save the frogs]

3 hours ago, scottiejohn said:

Correct!

forgetting about code for the time being.

this thing has the ability to create original poetry.

i showed an example in the other thread.

this means it has some form of creative intelligence.

most human beings cannot write a decent poem. 

 

[ukrules]

4 hours ago, OneMoreFarang said:

ChatGPT writes programming code which it didn't just copy and paste from somewhere. Try it! Just tell it in plain English what the program should do and which programming language you prefer. And then watch! 

Yes, it's useful when you're using a language you don't use all the time, it's quicker than reading the reference guides or going to a various stackoverflow / stackexchange type pages to get an example of what you want.

 

I took out a subscription to it as I do find it useful at times.

 

[save the frogs]

2 hours ago, rose33 said:

If you can't be bothered to truly test  ChatGPT's abilities

looks like i just been sKooled.

[OneMoreFarang]

3 hours ago, rose33 said:

Seems to me this big change is coming, whether any individual likes it or not.

 

Whether anyone personally ignores ChatGPT or enthusiastically throws  questions at it (intelligent or unintelligent) every evening, won't make a scrap of difference.

 

Some people just like to jump onto every bandwagon.

 

I know a very high IQ medical doctor, who has published extensively and completed a Fellowship, and also done a lot of charity work, who tested negative to COVID four times in one week, and still insists ????????????that he had COVID at that time. "Based on what, Sir? A HUNCH???" (The doctor turns witchdoctor!)

Just a full on bandwagon rider. Has to share the same experiences that everyone else. Some people are just like that.

 

Makes no difference in the end -- so I think people should feel emboldened to "dare to ignore" (your words) ChatGPT and all the AI buzz, if that's their preference.

 

Nature is still there, instrumental music, animals, cooking... all the good things that were always there are still there. 

 

I for one cherish the opportunity to talk to those who still have the ability to generate THEIR OWN THOUGHTS. These days most people cannot converse; they merely regurgitate from news and social media. Like demented parrots. ChatGPT will only exacerbate the problem.

 

I should pick out a few choice individuals right now and beg them to ring fence with me ???? The Anti ChatGPT Chat Group.

 

Sure, people don't have to use it. They can ignore it. But I think it makes a lot of sense that people keep up to date about the capabilities from AI.

Imagine someone would ignore the internet. Sure, he/she doesn't have to use any of it personally. But it's there, almost everybody is using it, with tremendous consequences. Pretending it doesn't exist doesn't remove the consequences of the fact that it exists. 

[Jingthing]

2 hours ago, save the frogs said:

forgetting about code for the time being.

this thing has the ability to create original poetry.

i showed an example in the other thread.

this means it has some form of creative intelligence.

most human beings cannot write a decent poem. 

 

Yes, not a great example but I wondered how it would do with a poem about Pattaya gay go go bars in the style of Celine. Celine wrote in French.

 

Quote

In Pattaya, where the neon light glows A place where sensuous pleasures flow Where men seek men for a night of lust In Go-Go bars with a tempting thrust The music's loud, the drinks are strong And the dancers move to a sultry song Orchestrated in a curious flair That beckons all but the unaware The patrons' eyes, they never stray From the muscular form on display Watching them as they work the floor

About the legalization of cannabis in Thailand in the style of Evelyn Waugh.

Quote

From Bangkok's sprawl of neon lights and noise, Rises up a haze of sweet tobacco's joys. Once taboo, now legal, the herb is free, A nod to Thailand's culture and history. Tourists flock to sample this new delight, To savor its flavor and take flight, As locals watch with cautious pride, A change in tide they cannot hide. Yet with this freedom comes great cost, As moralists lament what's been lost, Tradition, decency, and social good

 

[save the frogs]

Question: What is a good pickup line for a girl in a bar in Soi 6 Pattaya in the style of The King James Bible?

 

ChatGPT:

"I sayeth unto thee, fair maiden, art thou an angel sent from the heavens above? For thou art more precious than rubies and thy beauty radiates brighter than the morning sun."

 

 

[rose33]

2 hours ago, OneMoreFarang said:

Sure, people don't have to use it. They can ignore it. But I think it makes a lot of sense that people keep up to date about the capabilities from AI.

Imagine someone would ignore the internet. Sure, he/she doesn't have to use any of it personally. But it's there, almost everybody is using it, with tremendous consequences. Pretending it doesn't exist doesn't remove the consequences of the fact that it exists. 

???? Great point - I agree there's where the focus should be. 

 

It's a bit like, along comes this new thing called the Model T Ford and, we're very excited, there's a lot of hype, so we focus on its speed, engine power and handling rather than trying to project into the future and guess at the longterm effects of mass use of automobiles. (Direct and indirect consequences). 

 

Or, along comes feminism, divorce law, and the sexual revolution, and we focus on the quality of our pick-up lines and on the relative efficacy of various contraceptives rather than imagining the likely longterm effects of redesigning the relationship between men & women. (Direct and indirect consequences). 

 

With AI, there's a preliminary question as to whether it does in fact exist (the topic of this thread).

 

If AI does exist as reported, what are the likely consequences for humanity? (Direct and indirect consequences). 

 

On the other hand, if AI doesn't exist as presented, and is only a lie being told by certain journalists, academics and netizens, then.... <drumroll>.... we need to ask, what might be the long-term, direct and indirect consequences for humanity?

What is the usual long-term result when we blindly trust invisible authority figures who excel in manipulating our emotions?

 

 

 

[rose33]

2 hours ago, save the frogs said:

Question: What is a good pickup line for a girl in a bar in Soi 6 Pattaya in the style of The King James Bible?

 

ChatGPT:

"I sayeth unto thee, fair maiden, art thou an angel sent from the heavens above? For thou art more precious than rubies and thy beauty radiates brighter than the morning sun."

 

 

Is it necessary to sweet-talk these women? I thought the point was to pay them...

 

"mai poot ploy, khun jaai ploy"

[Dan O]

8 hours ago, rose33 said:

 

IQ is not the same as wisdom, and IQ is not a predictor of a person's usefulness to others  High-IQ people don't necessarily lead productive lives, nor necessarily bring peace and happiness to the people & communities around them.

 

High-IQ people tend to feel lonely and misunderstood. They often spend time alone because they prefer that over human company. They can become rather bitter and resentful people. When they do engage, it is comparably easy for them to exploit and manipulate others -- not because they are evil people, but for the practical reason that their processing speed & accessible memory is greater.

 

All other things being equal, I wouldn't  place a lot of trust in someone purely because of their high IQ. I would need more information about what they are like as a person, so I could get an overall picture.

 

 

 

12 minutes ago, rose33 said:

???? Great point - I agree there's where the focus should be. 

 

It's a bit like, along comes this new thing called the Model T Ford and, we're very excited, there's a lot of hype, so we focus on its speed, engine power and handling rather than trying to project into the future and guess at the longterm effects of mass use of automobiles. (Direct and indirect consequences). 

 

Or, along comes feminism, divorce law, and the sexual revolution, and we focus on the quality of our pick-up lines and on the relative efficacy of various contraceptives rather than imagining the likely longterm effects of redesigning the relationship between men & women. (Direct and indirect consequences). 

 

With AI, there's a preliminary question as to whether it does in fact exist (the topic of this thread).

 

If AI does exist as reported, what are the likely consequences for humanity? (Direct and indirect consequences). 

 

On the other hand, if AI doesn't exist as presented, and is only a lie being told by certain journalists, academics and netizens, then.... <drumroll>.... we need to ask, what might be the long-term, direct and indirect consequences for humanity?

What is the usual long-term result when we blindly trust invisible authority figures who excel in manipulating our emotions?

 

 

 

 

6 hours ago, rose33 said:

Seems to me this big change is coming, whether any individual likes it or not.

 

Whether anyone personally ignores ChatGPT or enthusiastically throws  questions at it (intelligent or unintelligent) every evening, won't make a scrap of difference.

 

Some people just like to jump onto every bandwagon.

 

I know a very high IQ medical doctor, who has published extensively and completed a Fellowship, and also done a lot of charity work, who tested negative to COVID four times in one week, and still insists ????????????that he had COVID at that time. "Based on what, Sir? A HUNCH???" (The doctor turns witchdoctor!)

Just a full on bandwagon rider. Has to share the same experiences that everyone else. Some people are just like that.

 

Makes no difference in the end -- so I think people should feel emboldened to "dare to ignore" (your words) ChatGPT and all the AI buzz, if that's their preference.

 

Nature is still there, instrumental music, animals, cooking... all the good things that were always there are still there. 

 

I for one cherish the opportunity to talk to those who still have the ability to generate THEIR OWN THOUGHTS. These days most people cannot converse; they merely regurgitate from news and social media. Like demented parrots. ChatGPT will only exacerbate the problem.

 

I should pick out a few choice individuals right now and beg them to ring fence with me ???? The Anti ChatGPT Chat Group.

 

Seems your stringing together some very vague concepts and ideas that are not really central to the AI issue raised.   CHATGPT is a small sliver of the AI universe of development and is a slim section of it. There are developments in AI that aren't in the main stream currently that are light years ahead of the playground present in CHATGPT. Your focus should really be within that arena if you truely want to understand the inpact AI can have with all it's good and ill aspects 

[save the frogs]

24 minutes ago, rose33 said:

Is it necessary to sweet-talk these women?

no. no game needed.

another useless chatGPT query. 

[Red Phoenix]

7 hours ago, rose33 said:

Seems to me this big change is coming, whether any individual likes it or not.

 

Whether anyone personally ignores ChatGPT or enthusiastically throws  questions at it (intelligent or unintelligent) every evening, won't make a scrap of difference.

 

Some people just like to jump onto every bandwagon.

 

I know a very high IQ medical doctor, who has published extensively and completed a Fellowship, and also done a lot of charity work, who tested negative to COVID four times in one week, and still insists ????????????that he had COVID at that time. "Based on what, Sir? A HUNCH???" (The doctor turns witchdoctor!)

Just a full on bandwagon rider. Has to share the same experiences that everyone else. Some people are just like that.

 

Makes no difference in the end -- so I think people should feel emboldened to "dare to ignore" (your words) ChatGPT and all the AI buzz, if that's their preference.

 

Nature is still there, instrumental music, animals, cooking... all the good things that were always there are still there. 

 

I for one cherish the opportunity to talk to those who still have the ability to generate THEIR OWN THOUGHTS. These days most people cannot converse; they merely regurgitate from news and social media. Like demented parrots. ChatGPT will only exacerbate the problem.

 

I should pick out a few choice individuals right now and beg them to ring fence with me ???? The Anti ChatGPT Chat Group.

 

From the article:

Members will soon be able to change preferences to auto-answer topics with realistic, automated replies that consider previous posting history. Member settings allow for automatic posting of replies, freeing up time when not at the keyboard.
In the future, moderation actions will also be automated.

= = =

Yep, a Forum where you won't know whether the replies on your post were auto-generated and you might unknowingly end up discussing with a machine.
And of course AseanNow introducing that novelty also want to use it for moderation, which would absolutely kill any deviation from the accepted narrative on controversial subjects.

The Ultimate Echo-Chamber...

[rose33]

3 minutes ago, Dan O said:

 

You say I am being vague because I'm not talking about the concrete thing immediately in front of you.

 

You think I am 'stringing together' concepts because I have a large data set in my head (I read a lot) and therefore I'm able to draw broader comparisons, which add to understanding and therefore lead us toward better quality answers.

 

This is a thread about AI generally, and whether it exists as hyped. Mainstream media presents ChatGPT as having made leaps and bounds in its development in recent months. These reported recent developments in ChatGPT are prompting more discussion about AI in the general population. Hence the focus on it in this thread in April 2023.

[Red Phoenix]

Here the 'intelligent' response from AseanNow's ChatGTP on my question.

 

[Denim]

If man can imagine something, he will eventually create that something. It's just a matter of time , money and a lot of work.

 

Look where mankind was 150 years ago and where it is today.

 

Big change , but not all good.

[rose33]

1 minute ago, Red Phoenix said:

Here the 'intelligent' response from AseanNow's ChatGTP on my question.

 

Prior to covid, my understanding of vaccination was that it prevented disease transmission (it prevented the body from 'catching' the relevant illness).

 

Prior to covid, i never heard of any vaccine that could reduce symptoms. 

 

Is there a gap in my knowledge, or did the basic function of vaccines suddenly change significantly in March 2020?