chatgpt

[L.W. Baxter]

just reading that the chatbot app is taking off faster than any app ever.

some suggestion that it could be a "threat to google".

i haven't used it and frankly don't want to. i rebel against the existence of this technology.

the shine is off "the information age" for me, as the potential for increased access to knowledge has given way to the grim reality: we have been tampered with.

stop the world i want to get off.

[Paul Pless]

stop the world i want to get off.

step one: trade your iphone for a flip phone

[Paul Pless]

step two

[L.W. Baxter]

it ain't all waitin' on me

[Garret]

Being a software guy & talking/interacting with others of the ilk - I'm concerned. Many see it as a logical step & I'm not at all sure I do. I may be paranoid, but I really don't need something with who knows what internal directives writing stuff for me. I don't want Alexa spying on me either.

The information age has become, in large part, the misinformation age. Has it helped connect people? Sure, look at our forum. Is knowledge more readily available? Sure again - but it's often buried under piles of garbage that many are not capable of sifting. The worst part IMO is the constant sharing of unvetted "information" that is passed off as "fact".

Think how AI (which is misnamed - ChatGPT is not "intelligent" - it's just capable of learning new stuff) can help spread whatever its creators want spread. Also remember this is early days & it will get far more sophisticated.

It's not just a threat to google - it's a threat to most everything. Need your resume written? No problem. A term paper? Sure - on what topic? A speech? Easy-peasy. Best way to handle an HR problem? Also easy. Best way to get elected? Best way to... It goes on & on & no one except the folks who put it together have any idea what is going on inside the system.

Chilling IMO

[bob winter]

I am glad I am pushing 80, the future doesn’t look all that appealing.I think I will give it a miss.

[bluedog225]

If I weren’t too cheap to spend the 9 bucks, I’d generate a chatgpt response to your post.

[L.W. Baxter]

i think there is a real possibility of becoming enslaved. voluntarily.

like filters that give everyone flawless skin. at what point does the projection become the dominant being?

if we surrender our communications to a machine with infinite "knowledge" and perfect diction, at what point does the human mind become a parasite.

i have been concerned that we will vote away our right to vote, uncertain that we can resist the temptation to abdicate. but the problem may run much deeper than age old cycles of power.

[Garret]

It's already writing computer code. Not good code, or even code that'll work well, but it's really early days. Think how great it'll be when the "AI" machines start writing "AI" machines. [end sarcasm]

[Joe (SoCal)]

Here the thing AI hasn't invented anything or has come up with an original idea. Sure with Chat GPT you can say write me a story about a wale ship adventure with a great white wale in the style of Herman Melville and you will get something similar to Moby Dick. Tell Chat GPT to make a painting of Sunflowers in the style of van Gogh, and you will get something similar to a van Gogh. Sure it's a nice parlor trick and will make people think it's totally sentient, when actually it just re-digesting what is already out there. It's not creating new paradigms, inventing new forms of art, music, science and literature it's just distilling what humans have brought to the table and shuffling the deck so it looks new.

As for educators concerned about cheating then teachers and the students have lost the idea that education is not about grades but about LEARNING and expanding and hopefully developing new and original concepts, inventing new forms of art, music, science and literature THAT HAVE NEVER BEFORE SEEN THE LIGHT OF DAY. Thats what Humans do and what AI can not do . .....................yet

[WI-Tom]

Teachers are watching this very closely, I can tell you. There's no putting the genie back into the bottle (though with this genie, I sure would if I could), so we have to deal with it.

It can definitely generate text that is better than the average high school student.

Seems like the logical (and horrible) next step in the long trajectory of dumbing down humanity by offloading essential intellectual/skill tasks to machines so we can... What? Vegetate, I guess.

Tom

[TomF]

It occurs to me that there have been not a small number of bits of literature and film devoted to essentially this story arc. They don't end well. Any of them.

What we tend to forget is that stories set in sci-fi futures (or costume-drama pasts) are never really about the alleged time period, they're about the present in which the authors live. The concerns, fears, problematic structures and relationships of their own times, which can be examined more minutely by putting them into an exotic setting.

What we haven't learned is that the capacity or drive to do something as an accomplishment doesn't mean that it ought to be done. Often, it ought not be. Such stories are re-tellings of how our species has a genius for ignoring our hunches and intuitions, when we think our immediate comfort or advantage could be helped.

AI is little different, except that if the materialist view of the conscious mind is true, we'll ultimately create one - at which point all bets are off. Self interest looks different to a bear, lion, or ungulate than it does to a human - why should we imagine an AI being would perceive its self interest in a way that fundamentally defaults to compassion for and deference to humans? Because we wrote it into the code? A code which the AI being itself could - and is designed to - rewrite?

[Garret]

Want to get really depressed about ChatGPT & education? https://www.npr.org/2023/01/26/11514...wharton-school

I see his point but...

[Bflat]

You can't stop progress, but you can unplug a lot of it.

[L.W. Baxter]

Here the thing AI hasn't invented anything or has come up with an original idea. Sure with Chat GPT you can say write me a story about a wale ship adventure with a great white wale in the style of Herman Melville and you will get something similar to Moby Dick. Tell Chat GPT to make a painting of Sunflowers in the style of van Gogh, and you will get something similar to a van Gogh. Sure it's a nice parlor trick and will make people think it's totally sentient, when actually it just re-digesting what is already out there. It's not creating new paradigms, inventing new forms of art, music, science and literature it's just distilling what humans have brought to the table and shuffling the deck so it looks new.

As for educators concerned about cheating then teachers and the students have lost the idea that education is not about grades but about LEARNING and expanding and hopefully developing new and original concepts, inventing new forms of art, music, science and literature THAT HAVE NEVER BEFORE SEEN THE LIGHT OF DAY. Thats what Humans do and what AI can not do . .....................yet

oh, i dunno.

haven't you recently come to the realization that your iphone was doing significant aspects of your photographing "art" for you?

and now you have found another camera that makes you feel like you have more control. and you are doing your best to recreate what the iphone was doing.

i hope you won't be too hurt if i ask, who is the artist here? whose creativity dominates? at what point is a photographer a remora hitched to a shark that eats the scenery.

[L.W. Baxter]

It occurs to me that there have been not a small number of bits of literature and film devoted to essentially this story arc. They don't end well. Any of them.

What we tend to forget is that stories set in sci-fi futures (or costume-drama pasts) are never really about the alleged time period, they're about the present in which the authors live. The concerns, fears, problematic structures and relationships of their own times, which can be examined more minutely by putting them into an exotic setting.

What we haven't learned is that the capacity or drive to do something as an accomplishment doesn't mean that it ought to be done. Often, it ought not be. Such stories are re-tellings of how our species has a genius for ignoring our hunches and intuitions, when we think our immediate comfort or advantage could be helped.

AI is little different, except that if the materialist view of the conscious mind is true, we'll ultimately create one - at which point all bets are off. Self interest looks different to a bear, lion, or ungulate than it does to a human - why should we imagine an AI being would perceive its self interest in a way that fundamentally defaults to compassion for and deference to humans? Because we wrote it into the code? A code which the AI being itself could - and is designed to - rewrite?

the concern about self-aware, and possibly hostile ai is another level entirely.

my first concern is for the changing nature of humans and our society, not computer intelligence.

it may be that by the time our android overlords are fully realized and self-replicating, we will welcome them with credit cards. maybe not even knowing what we have given up. and at the moment, i fear this happening in my lifetime. the process is already under way.

[WI-Tom]

It occurs to me that there have been not a small number of bits of literature and film devoted to essentially this story arc. They don't end well. Any of them.

What we tend to forget is that stories set in sci-fi futures (or costume-drama pasts) are never really about the alleged time period, they're about the present in which the authors live. The concerns, fears, problematic structures and relationships of their own times, which can be examined more minutely by putting them into an exotic setting.

What we haven't learned is that the capacity or drive to do something as an accomplishment doesn't mean that it ought to be done. Often, it ought not be. Such stories are re-tellings of how our species has a genius for ignoring our hunches and intuitions, when we think our immediate comfort or advantage could be helped.

AI is little different, except that if the materialist view of the conscious mind is true, we'll ultimately create one - at which point all bets are off. Self interest looks different to a bear, lion, or ungulate than it does to a human - why should we imagine an AI being would perceive its self interest in a way that fundamentally defaults to compassion for and deference to humans? Because we wrote it into the code? A code which the AI being itself could - and is designed to - rewrite?

"Open the pod bay doors, Hal."

"I'm sorry, Dave. I'm afraid I can't do that."

Tom

[Joe (SoCal)]

Years ago in my late teens I was obsessed the emerging world of computers. I read a lot of Gödel, Escher, Bach: an Eternal Golden Braid by Douglas Hofstadter, The Origin Of Consciousness In The Breakdown Of The Bicameral Mind by Julian Jaynes. I was playing on an old mainframe that had a copy of ELIZA ( ELIZA is an early natural language processing computer program created from 1964 to 1966 at the MIT Artificial Intelligence Laboratory by Joseph Weizenbaum.) It was a simple type / response program that used Rogerian psychotherapy Person-centered therapy seeks to facilitate a client's self-actualizing tendency, "an inbuilt proclivity toward growth and fulfillment", via acceptance (unconditional positive regard), therapist congruence (genuineness), and empathic understanding. Oh and man did it seem REAL, some people believed it could pass the Turing Test ( it couldn't ) until you distilled that it was just re-asking you the same things you were telling it. Again it was just another parlor trick.

https://en.wikipedia.org/wiki/Person-centered_therapy

You wan't to see machines that give me the creeps just go look at Boson Dynamics videos

[WI-Tom]

the concern about self-aware, and possibly hostile ai is another level entirely.

my first concern is for the changing nature of humans and our society, not computer intelligence.

it may be that by the time our android overlords are fully realized and self-replicating, we will welcome them with credit cards. maybe not even knowing what we have given up. and at the moment, i fear this happening in my lifetime. the process is already under way.

Mine, too. How eagerly, how eagerly, we rush to hand over the "drudgery" of doing things for ourselves to our machines. At which point, when we no longer have to do anything, we will...

??? What's the point? What's the attraction?

But there is a strong thread of technological fundamentalism in our culture, where to question tech is seen as bad, reactionary, etc.

I'm a little surprised to see that some other WBF members have come around to recognizing what I've often posted about--our growing reliance on tech, and how it's a terrible thing in many ways. I used to feel like a lonely voice when I raised those concerns.

Tom

[L.W. Baxter]

to illustrate what i mean by "already under way", consider online dating.

the trend these days, according to a young man of my aqcuaintance, is for profile pictures that have been so filtered that they might as well be avatars. anybody who posts a realistic portrait is out of the game. and avatars meeting in real life tend to be disappointed.

with the development of virtual reality, at some point, sex between avatars may become generally preferred (if it isn't already for some).

at that point, human nature is fundamentally altered. so many of our motivations spring from our drive for mating, companionship, maintanence of relationships. what will be left of us when everyone is wearing a headset and masturbating between bouts of feeding, sleeping, and streaming ai produced stimulation.

[WI-Tom]

Baxter, I think you're talking about something like this:



It's all already happening, I'd say. Online life and social media is "real" now--everything else is irrelevant, or quickly becoming so.

Tom

[Garret]

Mine, too. How eagerly, how eagerly, we rush to hand over the "drudgery" of doing things for ourselves to our machines. At which point, when we no longer have to do anything, we will...

??? What's the point? What's the attraction?

But there is a strong thread of technological fundamentalism in our culture, where to question tech is seen as bad, reactionary, etc.

I'm a little surprised to see that some other WBF members have come around to recognizing what I've often posted about--our growing reliance on tech, and how it's a terrible thing in many ways. I used to feel like a lonely voice when I raised those concerns.

Tom

Even though I'm in the IT biz, I've often wondered if it really is a good thing. Oh sure, it removes a lot of drudgery in the day-to-day lives of the corporate world, but at what cost? Big data being collected & mined by corporations is a great example. The amount of info that is known & shared about people is scary & privacy is at best an illusion. What surprises me the most is how people gleefully share private info online - never thinking about how it will be used down the road.

When all that data is accessed by "AI" & the "AI" programs start really utilizing it, what will be the end results? I don't think anyone knows.

As said above - just because it can be done doesn't mean it should.

[TomF]

to illustrate what i mean by "already under way", consider online dating.

the trend these days, according to a young man of my aqcuaintance, is for profile pictures that have been so filtered that they might as well be avatars. anybody who posts a realistic portrait is out of the game. and avatars meeting in real life tend to be disappointed.

with the development of virtual reality, at some point, sex between avatars may become generally preferred (if it isn't already for some).

at that point, human nature is fundamentally altered. so many of our motivations spring from our drive for mating, companionship, maintanence of relationships. what will be left of us when everyone is wearing a headset and masturbating between bouts of feeding, sleeping, and streaming ai produced stimulation.

"The Matrix" wasn't supposed to be aspirational.

[L.W. Baxter]

consider legions of avatars deciding to permanently quench maternal/paternal instincts with avatar babies.

and then the avatars and avatar babies continue the game after the death of the physical participant.

people have been in accidental training for this eventuality for years.

[bluedog225]

6797345[/URL]]oh, i dunno.

haven't you recently come to the realization that your iphone was doing significant aspects of your photographing "art" for you?

and now you have found another camera that makes you feel like you have more control. and you are doing your best to recreate what the iphone was doing.

i hope you won't be too hurt if i ask, who is the artist here? whose creativity dominates? at what point is a photographer a remora hitched to a shark that eats the scenery.

I don’t want a drill and a bit; I want a hole in something.

[bluedog225]

And people creating real art probably thought the same thing when cameras appeared.

[L.W. Baxter]

it's not unfair to point out that older people have long complained about change.

but i would counter that, objectively, the pace of change has been accelerating on an exponential curve. how far are we from a nearly vertical rate of change? change so fast that our culture will not adapt, but become an artifice imposed.

[C. Ross]

ChatGPT is fluent and charming but often wrong.

I think it will be used widely in places where those attributes are ok: customer service initial contact, simple tasks like reservations, media, entertainment, games. I don’t think companies are going to depend on it to make significant decisions and I don’t think consumers are going to trust it when they need to be sure an answer is correct.

I think it will dominate the Bilge - it’s talkative and often wrong. If it could be trained to be more talkative the more wrong it is, humans in the Bilge are doomed.

[L.W. Baxter]

imagine two chatbots meeting on a thread.

actually, i think we may have already had that.

[Garret]

ChatGPT is fluent and charming but often wrong.

I think it will be used widely in places where those attributes are ok: customer service initial contact, simple tasks like reservations, media, entertainment, games. I don’t think companies are going to depend on it to make significant decisions and I don’t think consumers are going to trust it when they need to be sure an answer is correct.

I think it will dominate the Bilge - it’s talkative and often wrong. If it could be trained to be more talkative the more wrong it is, humans in the Bilge are doomed.

It is often wrong - but remember this is early days & it (& all the others) will get better very quickly.

Wonder how many WBF logins are already "AI"?

[bluedog225]

What change? We eat, poop, bitch, and die.

[Dave Lesser]

If I weren’t too cheap to spend the 9 bucks, I’d generate a chatgpt response to your post.

ChatGPT is free

ChatGPT is fluent and charming but often wrong.

[TomF]

It is often wrong - but remember this is early days & it (& all the others) will get better very quickly.

Agreed. The principle is that the software's designed to reflect on it's experience, and learn. Because it's digital (rather than a lump of meat inside a skull), it has the great advantage of remembering fairly easily.

42.

[TomF]

Lee? Twodot?

You sure it isn't just your respective dogs that learned to type?

[mister fergus]

i always lick my— i mean log in on my own ba—i mean log in.