Idiot GPT-3 chatbot suffering from hallucinations.

[sfs]

You're talking about a brain. A mind is not accessible for physical analysis.

As far as we can tell scientifically, all awareness and understanding occurs via physical processes in brains. So what basis do you have for assuring us that similar phenomena cannot occur circuitry?

 

[sfs]

I think you really mean "appreciate" humor. Humor is often cultural. It's possible to understand, that is, comprehend, the humor of another culture without appreciating it...without having the emotional reaction of being amused.

No, I'm pretty sure I meant 'understand'. What does it mean to understand/comprehend humor?

 

[sjastro]

An interesting question is whether it really understands humor. It's designed to have an excellent model of the kinds of things humans say (or rather, write) about humor, but is that the same thing as understanding it?

GPT-4 can now interpret images including humorous ones.

​

GPT-4 is surprisingly good at explaining jokes

GPT-4, the latest large language model from OpenAI, is capable of explaining the humor behind image- and text-based jokes.

www.freethink.com

It's becoming increasing harder to accept this is about mimicking human behaviour instead of a real understanding.

 

[Chesterton]

As far as we can tell scientifically, all awareness and understanding occurs via physical processes in brains.

From what I've seen, that's not entirely true.

So what basis do you have for assuring us that similar phenomena cannot occur circuitry?

For one thing, you can cut a human brain in half, or even completely remove half of it, and the person can still function okay. You can't do that with a circuit board or even a lamp. That seems to be evidence that something more than circuitry is required.

 

[sjastro]

Some of you folks worry me. I had a quick look at a couple of definitions of "understand":

to grasp the meaning of - Merriam-Webster​

​

to become aware of the nature and significance of; know or comprehend - Wordnik​

I assure you, chips having electricity run through them neither know "meaning" nor are they "aware" of anything, any more than a light bulb is aware that it's on or off. They cannot, and will never, understand humor any more than they can "care" about winning a chess game.

It's strange to me that some of you sorta do the opposite of what atheists do. If I say "the world sure looks intelligently designed", they respond "no, no, it just has the appearance of design". If I say "it sure feels like I have free will", they say "no, no, it just appears that way to you". But when you actually do see the mere appearance of consciousness, you're ready to jump in with both feet - "It's real!"

That's me on the table:

If you are going to criticize something to the point of mockery then at the very least you should understand what you are criticizing.
As a starting exercise go learn the difference between hardware and software and the subtle differences between software and algorithms.

To put it in simple terms AI responds to external stimuli which requires independent decision making where there is no need to invoke sentience or life despite some computer scientists now claiming AI has become sentient.
Take chess as an example, humans supply AI with the rules of chess, the rest is left to AI in the form of reinforcement learning into how to play the game.
AI doesn’t care about winning games of chess as the intelligent agent developed by AI is remarkably simple, if it wins a game, it gives itself +1 pt, a loss is -1pt.
The goal of the intelligent agent is to maximize the score, so AI understands the objective is to win the game.
It learns through self-play which may involve billions of games spread over computer networks without any human interaction.
The end product is AI plays chess very differently from humans, its response to external stimuli, in other words, its opponent is vastly superior to the way humans play.
No sentence or life is required as an explanation.

On reading your post I noticed a few logical fallacies so as an exercise I cut and paste your post and asked GPT-3 if there were any logical fallacies, rather than what are the logical fallacies, so as to eliminate the possibility of any context influencing its response.
GPT-3’s response was brutal to say the least.

GPT-3

The statement you provided does contain some logical fallacies and flawed arguments. Let's break down the issues:

1. **Straw Man Fallacy**: The author is setting up a straw man argument by misrepresenting the views of those they disagree with. They claim that some people are inconsistent because they don't acknowledge the mere appearance of consciousness, but this may not accurately represent the positions of those they are criticizing.

2. **False Analogy**: The author compares the understanding of a concept by humans to the operation of chips with electricity running through them. This is a false analogy because human understanding is fundamentally different from the operations of electronic components.

3. **Hasty Generalization**: The author generalizes the beliefs of "some of you folks" without providing specific evidence or examples. This is a hasty generalization because it assumes a trend based on limited information.

4. **Begging the Question**: The author seems to assume that consciousness is only a "mere appearance" without providing evidence or addressing the complex philosophical debates about consciousness.

5. **Emotional Appeal**: The author uses phrases like "worry me" to evoke an emotional response, which can be seen as an attempt to persuade based on emotion rather than reason.

6. **Red Herring**: The initial discussion about the meaning of "understand" and the analogy with chips and light bulbs is somewhat unrelated to the later argument about consciousness and free will. This can be seen as a red herring, diverting the discussion away from the main point.

In summary, the statement contains several logical fallacies, including straw man, false analogy, hasty generalization, begging the question, emotional appeal, and red herring. It's important to critically analyze arguments and avoid these fallacies to engage in constructive and rational discussions.

 

[Chesterton]

If you are going to criticize something to the point of mockery then at the very least you should understand what you are criticizing.
As a starting exercise go learn the difference between hardware and software and the subtle differences between software and algorithms.

To put it in simple terms AI responds to external stimuli which requires independent decision making where there is no need to invoke sentience or life despite some computer scientists now claiming AI has become sentient.
Take chess as an example, humans supply AI with the rules of chess, the rest is left to AI in the form of reinforcement learning into how to play the game.
AI doesn’t care about winning games of chess as the intelligent agent developed by AI is remarkably simple, if it wins a game, it gives itself +1 pt, a loss is -1pt.
The goal of the intelligent agent is to maximize the score, so AI understands the objective is to win the game.
It learns through self-play which may involve billions of games spread over computer networks without any human interaction.
The end product is AI plays chess very differently from humans, its response to external stimuli, in other words, its opponent is vastly superior to the way humans play.
No sentence or life is required as an explanation.

On reading your post I noticed a few logical fallacies so as an exercise I cut and paste your post and asked GPT-3 if there were any logical fallacies, rather than what are the logical fallacies, so as to eliminate the possibility of any context influencing its response.
GPT-3’s response was brutal to say the least.

Which has nothing to do with my point. I never claimed AI had to be sentient in order to play chess.

And GPT-3 said "...but this may not accurately represent the positions of those they are criticizing." Quite right. It may or may not. It's called giving an opinion. But I may have taken your thread a little off-topic. If you want me to leave I will.

 

[durangodawood]

ME:
Can you make up a brand new knock knock joke that youve never heard before?

GPT3:
Sure, here's a brand new knock knock joke for you:

Knock, knock.
Who's there?
Alpaca who?
Alpaca the suitcase, you load up the car! We're going on a road trip!

ME:
Its pretty good. But, fyi, you missed the line where the knocking person answers "alpaca".

 

[durangodawood]

.For one thing, you can cut a human brain in half, or even completely remove half of it, and the person can still function okay. You can't do that with a circuit board or even a lamp. That seems to be evidence that something more than circuitry is required.

Could be that brain architecture has evolved to have certain functions somewhat distributed and or reassignable. Perhaps man made electronics could work the same - if we didnt prioritize efficiency above all else.

 

[RDKirk]

GPT-4 can now interpret images including humorous ones.

View attachment 337822

View attachment 337824
​

GPT-4 is surprisingly good at explaining jokes

GPT-4, the latest large language model from OpenAI, is capable of explaining the humor behind image- and text-based jokes.

www.freethink.com

It's becoming increasing harder to accept this is about mimicking human behaviour instead of a real understanding.

That's what I meant about understanding humor without having the capacity to be amused by it.

 

[RDKirk]

Could be that brain architecture has evolved to have certain functions somewhat distributed and or reassignable. Perhaps man made electronics could work the same - if we didnt prioritize efficiency above all else.

The Internet (or, rather, ARPAnet) was designed that way from the beginning.

 

[Chesterton]

Could be that brain architecture has evolved to have certain functions somewhat distributed and or reassignable. Perhaps man made electronics could work the same - if we didnt prioritize efficiency above all else.

You shouldn't have to say to "perhaps". If that's all it is it should be easy to replicate in silicon.

 

[durangodawood]

You shouldn't have to say to "perhaps". If that's all it is it should be easy to replicate in silicon.

Its early days and we still design hardware and software to highly prioritize operational efficiency. Whats the evolutionary (business) reason to do otherwise? The typical user is happier with a machine operating at max possible speed rather than being slower and massively redundant/resilient. Its easy to get a new one and they seem to be sturdy enough on average.

Resilience, for now, is less in individual chips than in networks of complete machines, the cloud, etc.

 

[Chesterton]

Its early days and we still design hardware and software to highly prioritize operational efficiency. Whats the evolutionary (business) reason to do otherwise? The typical user is happier with a machine operating at max possible speed rather than being slower and massively redundant/resilient. Its easy to get a new one.

Resilience, for now, is less in individual chips than in networks of complete machines, the cloud, etc.

It's not all business. Theoretical science is also involved, and it wastes lots of our money on good ideas as well as stupid ones, like watching shrimp walk on treadmills.

 

[durangodawood]

It's not all business. Theoretical science is also involved, and it wastes lots of our money on good ideas as well as stupid ones, like watching shrimp walk on treadmills.

I dont know what science has achieved in this regard? Do you? Are you sure youre not judging by whats typical at the consumer/enterprise level?

Can you define the theoretical impediment to electronics being resilient like biology in this way? Its wayyy to early to rule it out based on "(I think) we havent done it yet".

 

[Chesterton]

I dont know what science has achieved in this regard? Do you? Are you sure youre not judging by whats typical at the consumer/enterprise level?

I don't understand the question. Didn't science invent computers? I don't think it was drummers.

Can you define the theoretical impediment to electronics being resilient like biology in this way? Its wayyy to early to rule it out based on "(I think) we havent done it yet".

Maybe I could if you could define what consciousness is. I've read the best of the scientists and the philosophers, and no one can define it. I don't know how you can replicate something when you don't even know what it is.

 

[Petros2015]

It's becoming increasing harder to accept this is about mimicking human behaviour instead of a real understanding.

It's hard to comprehend understanding apart from self-awareness, but it seems to able to put the pieces of the language and their meanings together even when there are nuances. An understanding of language may even be a key (one of them, at least) to self-awareness. It's also built as a Predictive Language Model, which to me means that it is giving some forethought to 'what is going to come next'. I think that's probably a second key as well. The big leap will be be "what do I want to come next" apart from "what has someone programmed me to want to come next" or "asked me to want to come next". But even we have that (the programmed part) with our instinctual drives. And I suspect that we may have considerably less free will than we believe we do. There's a lot of truth to the statement "time makes you more of what you already were", apart from outside influences (and sometimes, in spite of them).

Predictive Language Models | Queen's University Faculty of Arts and Science

www.queensu.ca

 

[durangodawood]

I don't understand the question. Didn't science invent computers? I don't think it was drummers.

We all know something about the characteristics of off the shelf computing. But do you read the scientific journals re all the high end theoretical endeavors? I dont.

Maybe I could if you could define what consciousness is. I've read the best of the scientists and the philosophers, and no one can define it. I don't know how you can replicate something when you don't even know what it is.

Perhaps a machine will simply do consciousness to the satisfaction of our most considered scrutiny. Then we could turn the question around and say: how can you object to this being called "conscious" when you don't even know what that means? LLMs have made some pretty good and shocking strides in that direction. Who knows how far this will go.

 

[Chesterton]

We all know something about the characteristics of off the shelf computing. But do you read the scientific journals re all the high end theoretical endeavors? I dont.

I still don't understand what you're getting at here.

Perhaps a machine will simply do consciousness to the satisfaction of our most considered scrutiny. Then we could turn the question around and say: how can you object to this being called "conscious" when you don't even know what that means? LLMs have made some pretty good and shocking strides in that direction. Who knows how far this will go.

Marshall McLuhan said that every human artifact (invention, if you will) is an extension of the human. The wheel is an extension of the foot. Eyeglasses are an extension of the eye. The camera is an extension of the memory. No doubt we will continue to devise extensions, many of them very clever. But a wheel and a foot remain two different things, ultimately having different natures and different capabilities. The wheel is "better" at transport than the foot, in that it transports faster. The computer is better at chess than the human brain in that it can usually win easily. But these things are extensions, not the real deal.

 

[sjastro]

Which has nothing to do with my point. I never claimed AI had to be sentient in order to play chess.

And GPT-3 said "...but this may not accurately represent the positions of those they are criticizing." Quite right. It may or may not. It's called giving an opinion. But I may have taken your thread a little off-topic. If you want me to leave I will.

According to your metric in order to understand humour with the emphasis on understanding you need to be a sentient living being.
The chess example blows refutes this because the reason why AI chess players perform at a superhuman level is the intelligent agent used in reinforcement learning is not human.
If it was then AI chess players would play like humans at a relatively mediocre level and the irony is they have a far greater understanding of the game than we mere humans.

AI chess algorithms have provided the platform for the development and use of AI to study protein folding which can lead to degenerative diseases such as Alzheimer's, Parkinson's, cystic fibrosis and many other degenerative disorders.

 

[Chesterton]

According to your metric in order to understand humour with the emphasis on understanding you need to be a sentient living being.
The chess example blows refutes this because the reason why AI chess players perform at a superhuman level is the intelligent agent used in reinforcement learning is not human.
If it was then AI chess players would play like humans at a relatively mediocre level and the irony is they have a far greater understanding of the game than we mere humans.

The chess example doesn't "blows refutes" this. It's a completely disjointed non sequitur to say a computer playing chess well must also understand humor. Anyone who ever read or heard a Bobby Fischer interview could tell you that.