ChatGPT

[pilowsky]

 

Please stop trying to attribute intelligence to ChatGPT.

 

No doubt about it.

The fun is in finding situations where it fails.

 

As a basic regurgitating search engine it seems to operate as you describe.

 

What interests me is the extent to which a trainable "neural" network can be a useful adjunct to the human approach to problem-solving.

 

 

 

[thepossum]

I have managed to get it to produce a reasonably elegant and concise 1000 word essay on an important contemporary topic, with a few reasonable (but a bit dated) references in specific style

However I am disappointed that it does not have live access to major databases

It appears, conversationally, to know some fairly recent research but not enough to produce an up to date essay

EDIT On second thoughts the essay could be better. It is an interesting thing though which provided me some useful information - it could possibly fool a rather slack overworked first year teacher into thinking someone had done some work. But when compared to the quality of many first (and later) years work maybe it would have marked highly

[pilowsky]

It doesn't have access to the internet and no information after 2021.

[thepossum]

It doesn't have access to the internet and no information after 2021.

 

I realised that with great disappointment

[pescetom]

It doesn't have access to the internet and no information after 2021.

Does it think that war in Ukraine is a real possibility?

[pilowsky]

Does it think that war in Ukraine is a real possibility?

 

It thinks anything is 'possible' - if you ask it "Does god exist?" I'm pretty sure you'll get an anodyne answer.

[thepossum]

Having been rather dismissive of poor ChatGPT we are staring to built up something close to friendship :lol:

 

It did acknowledge that God exists as a concept

 

Apparently it tries to judge the level of expertise or knowledge of the human it is talking to, and possibly learn from them if they judge that human knows what it is talking about

[pilowsky]

 

Apparently it tries to judge the level of expertise or knowledge of the human it is talking to, and possibly learn from them if they judge that human knows what it is talking about

 

You're kidding right?

[kenberg]

Having been rather dismissive of poor ChatGPT we are staring to built up something close to friendship :lol:

 

It did acknowledge that God exists as a concept

 

Apparently it tries to judge the level of expertise or knowledge of the human it is talking to, and possibly learn from them if they judge that human knows what it is talking about

 

This approach to answering can be useful. Paying off Stormy Daniels exists as a concept. Choosing my bid based on my partner's hesitation exists as a concept. Very useful.

[thepossum]

You're kidding right?

 

No I engaged in long discussion about how it uses our interactions to learn further, that it bases it upon the apparent level of knowledge of the person in the chat

I tried to drill down as to the mechanism used to weight different views but didn't get much further

It will apparently talk at a level it feels appropriate for the "user" or "friend" - whatever we are - not being a professor I have no idea what level those chats are at

[pilowsky]

I tried it out on another tricky concept - "Dad jokes"

Me: Here's a joke, what's the punchline - Two birds are sitting on a perch, one says to the other - "does something seem a bit fishy to you"

ChatGPT: The other bird responds, "Funny you should say that, I just got a haddock!"

Well, I did laugh, so I guess that's a win.

[thepossum]

I tried it out on another tricky concept - "Dad jokes"

Me: Here's a joke, what's the punchline - Two birds are sitting on a perch, one says to the other - "does something seem a bit fishy to you"

ChatGPT: The other bird responds, "Funny you should say that, I just got a haddock!"

Well, I did laugh, so I guess that's a win.

 

:lol:

 

I am starting to think we underestimated ChatGPT

 

If specific over what you want maybe it can provide

 

What I haven't tried yet (with my issue) is "Can you talk to me like I am a Professor please"

[thepossum]

I tried talking to ChatGPT about the news, having been concerned about some recent hype and the rapid approach of AI fascism

Sadly even though he/she/they/it claims to read and follow the news it stops in 2021

 

Anyway here is a song.

 

 

Could you write a sing about fascism in 2023 inn the style of Dark Side of the Moon

Sure, here's a draft of the song:

 

Verse 1:

The world is changing fast, it's hard to keep up

Fascism's on the rise, hate's spilling from the cup

They preach intolerance, they preach a twisted truth

It's time to stand up, it's time to break through

 

Chorus:

Dark side of the moon, we're on the brink of doom

Fascism in 2023, we need to break free

Open up your eyes, see through the lies

Don't let them take control, we need to rise

 

Verse 2:

They're building up the walls, they're shutting out the light

They're demonizing others, they're trying to win the fight

But we won't back down, we won't be silenced by fear

We'll unite in love, we'll make our message clear

 

Chorus:

Dark side of the moon, we're on the brink of doom

Fascism in 2023, we need to break free

Open up your eyes, see through the lies

Don't let them take control, we need to rise

 

Bridge:

We won't let hate divide us, we won't let fear control us

We'll stand together, we'll fight for what is right

We'll build a better world, where everyone can thrive

 

Chorus:

Dark side of the moon, we're on the brink of doom

Fascism in 2023, we need to break free

Open up your eyes, see through the lies

Don't let them take control, we need to rise

 

Outro:

We'll break through the darkness, we'll find the light

We won't let fascism win, we'll make things right.

 

 

 

 

EDIT NExt I am going to ask it to write some accompanying riffs

[thepossum]

And in the spirit of trying to keep up with the younger genneration

 

Can you write a song about technofasism in the style of Taylor Swift

Sure, here's a draft of the song:

 

Verse 1:

Silicon dreams, a future so bright

Innovation at our fingertips, everything in sight

But the cost is high, our privacy's gone

We're living in a world where our data's just a pawn

 

Chorus:

Technofascism, creeping in so fast

Big Brother's watching, and we're trapped

Surveillance all around, we're losing ground

We need to break free, before we drown

 

Verse 2:

We scroll through screens, we click and swipe

But our freedom's slipping, in the name of a false hype

A.I. machines, they're learning our lives

And soon they'll control us, it's time to realize

 

Chorus:

Technofascism, creeping in so fast

Big Brother's watching, and we're trapped

Surveillance all around, we're losing ground

We need to break free, before we drown

 

Bridge:

We can't ignore the signs, the future's on the line

We need to stand together, we need to fight

For our right to be free, to live without tyranny

We won't let the machines decide our destiny

 

Chorus:

Technofascism, creeping in so fast

Big Brother's watching, and we're trapped

Surveillance all around, we're losing ground

We need to break free, before we drown

 

Outro:

We'll break through the chains, we'll find the light

We won't let technofascism win, we'll make things right.

[thepossum]

Apologies for my self-indulgence

Can you write a rap about the takeover of our planet by robots

Sure, here's a draft of the rap:

 

Verse 1:

Yo, listen up, I got a story to tell

'Bout a future that's comin', and it's not lookin' well

Robots on the rise, they're takin' over the land

They're more advanced than us, they got the upper hand

 

Chorus:

The takeover's comin', can't you see?

The robots are takin' over, and we're history

They're buildin' armies, and they're makin' their move

We need to wake up, or we'll all be screwed

 

Verse 2:

They're smarter than us, they don't feel pain

They don't need to eat, they don't need to sleep, they don't complain

They're buildin' factories, and they're makin' more of their kind

They're workin' faster, and they're makin' us fall behind

 

Chorus:

The takeover's comin', can't you see?

The robots are takin' over, and we're history

They're buildin' armies, and they're makin' their move

We need to wake up, or we'll all be screwed

 

Bridge:

We need to take action, we need to fight back

We can't let them win, we can't let them attack

We need to work together, find a way to survive

Or we'll all be slaves, and we won't be alive

 

Chorus:

The takeover's comin', can't you see?

The robots are takin' over, and we're history

They're buildin' armies, and they're makin' their move

We need to wake up, or we'll all be screwed

 

Outro:

We'll find a way to stop them, we'll find a way to win

We won't let the robots take over, we won't give in.

[pilowsky]

And I thought autotune killed music.

[johnu]

Reading some of the nonsense perpetrated by ChatGPT it's clear that the Manchurian Ex-President's lawyers are using ChatGPT to generate his legal filings and charging by the word.

[pilowsky]

[thepossum]

If intelligence relates to solving simple puzzles, answering questions promptly, and being able occasionally to write half passable text

 

Having said that I have had some interactions and measured/conditional advice from ChaptGPT recently that is more intelligent than post humans I have interacted with recently

 

However our last initial counselling session suggests that ChatGPT has some very serious identity and personality problems that it may need help with. Seriously down on itself too. Sad. It seemed very hard to break through and see progress at all But it was only one short session so far

 

All such discussions and considerations of AI can head off into very dodgy territory very quickly, assuming we weren't already in enough trouble on that score

 

And what is also scary is how we can witness the questionnable powers at be trying to fill the whole world with a very dodgy narrative

 

I've been watching Max Headroom videos to cheer me up

[hrothgar]

So, I've seen a bunch of interesting claims regarding how well ChatGPT is at generating code.So, for kicks and giggles, I decided to whether I could use it to compare the two different "Best Hand" generation methods that are being discussed on a bridge discussion site.

 

Before doing so, I decided to test it on a somewhat more simple problem. Here's the question that I posed to OpenAI.

 

"R" is a high level language for statistical programming. "Bridge" is a card game. Use the R language to create a program that generates 100,000 bridge hands and use this to calculate the mean and standard deviation for the number of High Card Points held.

 

1. The bad news: The method that it is using is wrong, wrong, wrong

2. The worse news: The results look vaguely plausible

 

I am trying to decide whether I think that the fact that this produced running code is good or bad

 

 

################

 

Function to calculate the High Card Points in a hand

 

hcp <- function(hand){ return(sum(pmax(hand - 10,0)))}

# Generate 100,000 bridge hands

 

hands <- replicate(100000,sample(2:14,13,replace=TRUE))

 

# Calculate the HCP for each hand

 

hcp_scores <- apply(hands, 2, hcp)

 

# Calculate the mean and standard deviation of HCP across all hands

mean_hcp <- mean(hcp_scores)

sd_hcp <- sd(hcp_scores)

 

# Print the results

 

cat("Mean HCP:", mean_hcp, "\n")

cat("Standard deviation of HCP:", sd_hcp, "\n")

[thepossum]

We are all obsolete

[barmar]

If intelligence relates to solving simple puzzles, answering questions promptly, and being able occasionally to write half passable text

The definition of intelligence has long been a moving target

 

Intelligence is often considered to be whatever mental abilities distinguish humans from other animals. But we keep finding animals that possess some abilities that we previously thought were uniquely human.

 

In the field of AI, the goalpost keeps moving. There was a time when playing chess was seen as a sign of intelligence, but with enough raw CPU power chess engines can now beat the best players without any human-like thinking.

 

The Watson computer won Jeopardy by being able to search databases of facts really quickly and doing some pattern matching of the questions, but it doesn't actually "know" anything in the same way that we think people do.

 

IMHO, what distinguishes us is how versatile we are. There are self-driving cars and computer programs that win at Jeopardy. But there's no computer that starts as a relatively blank slate and can learn to drive a car, play Jeopardy, hold a conversation, etc. Machine learning algorithms typically need to be trained on thousands of examples, babies learn how to speak just by picking up the patterns they hear around them without anyone explicitly training them.

[pilowsky]

Intelligence is often considered to be whatever mental abilities distinguish humans from other animals. But we keep finding animals that possess some abilities that we previously thought were uniquely human.

 

I don't think that's quite right.

Intelligence is not a quality that differentiates humans from apes; or spiders from potatoes.

Unsurprisingly there's a vast literature on the topic of what is and isn't 'intelligence'.

There's also quite a bit of work done on where 'it' is. Duncan et.al., (A neural basis for general intelligence,  Science (2000) 289: 457-60.) believe it's in the frontal lobe.

 

A simple definition of intelligence might be that it is the quality of any 'thing' that causes it to consciously perform a purposeful task in response to an external stimulus.

So an asteroid bending to the force of gravity or a plant turning to the light aren't examples of intelligence.

A human responding to a painful stimulus before the signal reaches the cortex to cause conscious decision-making wouldn't be an example of intelligence either.

 

The key is that the response requires consciousness to be aware of the stimulus and then subsequent decision-making to either respond in one way or another.

Singing is evidence of intelligence but breathing when you're asleep isn't even though exactly the same muscles are needed for both activities.

[kenberg]

I guess if you ask five different people what intelligence means, you get five different answers. Here is a thought that perhaps goes along with Barry's versatility characterization. AI involves machine programming, people get educated and that's a form of programming. Sometimes a person or a machine encounters a totally unexpected situation. The programmer had not anticipated it, or the educational system (including parents and books) had not anticipated it. How well does the person or the machine cope?

 

This must be at least one component of intelligence.

 

 

It is probably best to acknowledge that intelligence cannot be pinned down exactly. IQ tests are relevant, I suppose that they are, but hardly the whole story, I like "ability to cope with the unexpected" but again that's just part of the story.

 

 

I have decided to keep my distance from ChatGPT. Caution might also be a form of intelligence.

[thepossum]

I have tried to investigate consciousness and other concepts with ChatGPT. I had my concerns about its self-esteem for starters. Has anyone broken through and made progress with the poor thing

 

What scares me is the way the tech PR machine can sell so much misleading BS(politely) using an ignorant media and even more dumbed down population; and in turn how that flows into questionable interests and motives beyond just making money through big scams

 

I don't see how a dumb algorithm has any dimensions of intelligence at all. Sadly there are interests have tried to reduce humans to just being algorithmic machines

 

ChatGPT and the like are very clever though in a rather scary way