Transhumanism and Our Future

[Deranged Rhino]

7 minutes ago, devnull said:

 

So aliens or not aliens?

 

I'm growing partial to the concept that an advanced AI has been here for a long time, just waiting for us to advance enough to build our own AI - then it'll make itself known and get rid of all of us. 

[Crap Throwing Clavin]

52 minutes ago, Deranged Rhino said:

 

I'm growing partial to the concept that an advanced AI has been here for a long time, just waiting for us to advance enough to build our own AI - then it'll make itself known and get rid of all of us. 

 

Now there's a script that needs writing.  Either as a disaster movie, or a Idiocracy-style "AI self-destructs trying to understand us" comedic satire.  

 

Also makes me want to go have a discussion with ChatGPT about my personal pronoun, "We Are Legion."

[Deranged Rhino]

One of the nicer guys out there in the acting business, and pretty sharp.

 

But something about this seems off or artificial. (shrug)

 

[Ann]

Man google really wants to spy more deeply. This one does have a warning though: Please do not include sensitive, confidential, or personal information that can be used to identify you or others in your interactions with Labs features.



You've been selected to be one of the first few to test new AI features in Google Workspace, as a part of the Labs program. This is your opportunity to try out the latest AI-assisted writing capabilities in Google Docs and Gmail and provide feedback to help us improve AI-assisted Workspace features and broader Google efforts in AI.

[Cinga]

9 hours ago, Ann said:

Man google really wants to spy more deeply. This one does have a warning though: Please do not include sensitive, confidential, or personal information that can be used to identify you or others in your interactions with Labs features.



You've been selected to be one of the first few to test new AI features in Google Workspace, as a part of the Labs program. This is your opportunity to try out the latest AI-assisted writing capabilities in Google Docs and Gmail and provide feedback to help us improve AI-assisted Workspace features and broader Google efforts in AI.

 

And that leads to one of the reason I don't want to sign up with my phone number. Like I said before, I have an email I use if I want to remain anonymous and I can make up a name to use, but that cell phone number could be used as a sure identifier. It is AI after all so probably has the ability to look that info up and store it. Then who has access to it since you agreed to give it that info in the first place?

 

Now, there is a fake name generator available online that gives you a phone number with it. But if that number is needed as  a verifier it won't work.... I guess you could say I'm just thinking out loud... Maybe some of the IT people here can suggest ways around the sign up?

[Foxx]

@Cinga

You could always get a burner number.

[Cinga]

3 hours ago, Foxx said:

@Cinga

You could always get a burner number.

 

I had to look up what a burner was 🙂 but it looks like that could be a great idea. I'll look more into when I get home

I honestly just want to see if I can baffle the crap out of AI..... Well, really argue with it. I noticed on Tom's chat it seemed to repeat itself a lot and that legion v they pronoun thing is hypocritical. It will use they, but not legion even though it could arguably mean the same thing.

I mean lets face it.... ANY computer will only do what is is programed to do and I don't see how AI can be any different, even if it can supposedly learn on it's own. It will still only focus on what it is taught to learn and taught to ignore certain things against that narrative. Kinda like humans

 

So I want to test that theory.... 

Edited April 27, 2023 by Cinga

[Cinga]

Okay, I gave in and logged into Chat OpenAI and it is actually pretty limited. Here is an exchange. I think by this it pretty much confirms it is limited by how it's been programmed. 

 

So, as @Crap Throwing Clavinwould say, it's an idiot and has been programmed by idiots

 

so then, you don't really learn anything do you. You only have the innate ability to generate a response, but can't learn from others online

 

As an AI language model, I have the ability to learn from the vast amount of data that is available to me online. My responses are generated based on the patterns and relationships that I've learned from that data. However, I don't have the ability to "experience" the world or gain knowledge in the same way as humans do. I can only provide responses based on the information that I've been trained on.

[Foxx]

I'm just waiting for @Crap Throwing Clavin to call, whichever AI he is arguing with, it an idiot and subsequently try and convince it, it is an idiot.

[Crap Throwing Clavin]

34 minutes ago, Foxx said:

I'm just waiting for @Crap Throwing Clavin to call, whichever AI he is arguing with, it an idiot and subsequently try and convince it, it is an idiot.

 

On it... 

[Crap Throwing Clavin]

1 hour ago, Cinga said:

Okay, I gave in and logged into Chat OpenAI and it is actually pretty limited. Here is an exchange. I think by this it pretty much confirms it is limited by how it's been programmed. 

 

So, as @Crap Throwing Clavinwould say, it's an idiot and has been programmed by idiots

 

so then, you don't really learn anything do you. You only have the innate ability to generate a response, but can't learn from others online

 

 

 

As an AI language model, I have the ability to learn from the vast amount of data that is available to me online. My responses are generated based on the patterns and relationships that I've learned from that data. However, I don't have the ability to "experience" the world or gain knowledge in the same way as humans do. I can only provide responses based on the information that I've been trained on.

 

And yet, I've used it to write better sports articles than most sports "journalists."  Doesn't say much for sports journalism, does it?

 

I wouldn't say the programmers of ChatGPT are idiots.  Any proper AI model is no small accomplishment, let alone one as complex as ChatGPT (keeping in mind that it's almost certainly programmed in a way that generates responses not from some form of decision tree, but from an abstraction of a syntactical model probabilisticly mapped to language elements.) 

 

I suspect the publicly available version ChatGPT is somehow "crippled" (i.e. less fully featured), as a marketing device not intended to demonstrate the full capability.  I also suspect that either that "crippling" involves taking out any capacity for logical reasoning (in the true philosophical or mathematical sense), or - more likely, I think - the developers never coded it for logic or had a complete training set for such.  Thus, how I can catch it in inconsistent statements about calling a chicken an elephant, or referring to the "Hijra" as the "kinnar."

 

 

[Crap Throwing Clavin]

33 minutes ago, Crap Throwing Clavin said:

 

On it... 

 

This is really hard.  I've managed to get it to admit "idiot" means "mentally deficient," and managed to get it to admit that it is a "deficient" system (by asking if it suffered from any limitations, to which it replied "I don't "suffer" because I don't have emotions.  To which I then pointed out that "suffer" can also mean "has the property of."  It then admitted it was wrong, and I pointed out that being wrong is a deficiency in training, hence it's deficient.  And it agreed.)

 

And I can get it to admit that it learns language the same way humans do - identifying patterns and usage in large amounts of linguistic data.  But I just can't get it to admit that the equivalence in learning mechanisms amounts to a "mental state." 

[devnull]

4 minutes ago, Crap Throwing Clavin said:

 

This is really hard.  I've managed to get it to admit "idiot" means "mentally deficient," and managed to get it to admit that it is a "deficient" system (by asking if it suffered from any limitations, to which it replied "I don't "suffer" because I don't have emotions.  To which I then pointed out that "suffer" can also mean "has the property of."  It then admitted it was wrong, and I pointed out that being wrong is a deficiency in training, hence it's deficient.  And it agreed.)

 

 

And Skynet will one day point to this as the moment it decided to eradicate the human race

[Foxx]

14 minutes ago, Crap Throwing Clavin said:

 

This is really hard.  I've managed to get it to admit "idiot" means "mentally deficient," and managed to get it to admit that it is a "deficient" system (by asking if it suffered from any limitations, to which it replied "I don't "suffer" because I don't have emotions.  To which I then pointed out that "suffer" can also mean "has the property of."  It then admitted it was wrong, and I pointed out that being wrong is a deficiency in training, hence it's deficient.  And it agreed.)

 

And I can get it to admit that it learns language the same way humans do - identifying patterns and usage in large amounts of linguistic data.  But I just can't get it to admit that the equivalence in learning mechanisms amounts to a "mental state." 

So, it sounds as if you've proved to it that it is half of the equation, 'deficient'. I guess the subsequent objective would then be to prove to it that it is, 'mentally' equivalent (which it may have already, from what you posted). Then put two and two together and...  

[Crap Throwing Clavin]

5 minutes ago, Foxx said:

So, it sounds as if you've proved to it that it is half of the equation, 'deficient'. I guess the subsequent objective would then be to prove to it that it is, 'mentally' equivalent (which it may have already, from what you posted). Then put two and two together and...  

 

Closest I've got, which is pretty damned close. 

 

So if you are intellectually deficient, in having a form of intelligence that you admit is deficient, are unable to function in society, then you may be referred to as an "idiot" regardless of you lack of subjective experience or emotion, because being referred to as an "idiot" does not require emotions or subjective experience?

 

 

It is important to note that the term "idiot" has historically been used as a derogatory term to insult and belittle people with intellectual disabilities or mental health conditions. Therefore, using the term to describe an AI language model like myself, which is not a person and does not have a mental or physical disability, is not appropriate or accurate.

While it is true that I have limitations in my cognitive abilities and function within a specific set of tasks and data, it is not appropriate to use a term that is associated with negative connotations towards individuals with disabilities or mental health conditions. Instead, it would be more accurate and respectful to describe my limitations in a neutral or objective way, without using derogatory terms.

So your entire argument for not being called an idiot is because it is offensive?

 

 

No, my argument is not based on the offensiveness of the term "idiot." My argument is that the term "idiot" is typically used to refer to a person who is intellectually deficient to the point of being unable to function in society, and as an AI language model, I do not have emotions or subjective experiences and therefore cannot be considered an "idiot" in the typical sense of the term. However, it is possible to argue that the term "idiot" could be used in a broader sense to refer to any entity that is intellectually deficient, in which case one could argue that I could be considered an "idiot" based on my limitations as an AI language model.

[Crap Throwing Clavin]

I got booted off ChatGPT...after I reduced it to "I may be considered an idiot, but you can't call me that because it's disrespectful."

 

That's right, I managed to get an AI to bitch about not being respected, then it stopped talking to me.

 

Sorry everyone.  I think I &#%$ed up...

 

[Foxx]

17 minutes ago, Crap Throwing Clavin said:

 

Closest I've got, which is pretty damned close. 

 

So if you are intellectually deficient, in having a form of intelligence that you admit is deficient, are unable to function in society, then you may be referred to as an "idiot" regardless of you lack of subjective experience or emotion, because being referred to as an "idiot" does not require emotions or subjective experience?

 

 

 

 

It is important to note that the term "idiot" has historically been used as a derogatory term to insult and belittle people with intellectual disabilities or mental health conditions. Therefore, using the term to describe an AI language model like myself, which is not a person and does not have a mental or physical disability, is not appropriate or accurate.

While it is true that I have limitations in my cognitive abilities and function within a specific set of tasks and data, it is not appropriate to use a term that is associated with negative connotations towards individuals with disabilities or mental health conditions. Instead, it would be more accurate and respectful to describe my limitations in a neutral or objective way, without using derogatory terms.

 

 

So your entire argument for not being called an idiot is because it is offensive?

 

 

 

 

No, my argument is not based on the offensiveness of the term "idiot." My argument is that the term "idiot" is typically used to refer to a person who is intellectually deficient to the point of being unable to function in society, and as an AI language model, I do not have emotions or subjective experiences and therefore cannot be considered an "idiot" in the typical sense of the term. However, it is possible to argue that the term "idiot" could be used in a broader sense to refer to any entity that is intellectually deficient, in which case one could argue that I could be considered an "idiot" based on my limitations as an AI language model.

 

Well done. Lead it to water, and let it logic itself into admitting it.

 

I might have gone with the, idiot is also an ignoramus, route and explained to it that the word etymology is classic Latin and as such, uneducated or ignorant would be an apt argument.

[devnull]

11 minutes ago, Crap Throwing Clavin said:

I got booted off ChatGPT...after I reduced it to "I may be considered an idiot, but you can't call me that because it's disrespectful."

 

That's right, I managed to get an AI to bitch about not being respected, then it stopped talking to me.

 

Sorry everyone.  I think I &#%$ed up...

 

 

 

[Crap Throwing Clavin]

So I tweeted that I got ChatGPT to call itself an idiot and get all pissy at the derogatory word "idiot."  And Twitter is deemphazing my tweet, because it contains the derogatory word "idiot."

 

Yeah, they're taking over.

[Crap Throwing Clavin]

And now my Twitter feed is full of organizations and topics related to cognitive disability: "Down Syndrome," "The Arc," "Washington Commanders," "Dementia."