That Thread Written by ChatGPT

[The Grand Nudger]

-I do, though, enjoy thinking up questions and setups to test their guardrails. I wonder what it would have to say if you asked it to write a paper on the death toll and causes of the bowling green massacre? I know that the programs have massive utility for propaganda, since their source input is already whatever nuts are thinking and saying online. No need to expend human effort or talent wondering what might trigger a loon and how to package it. The loons tell the chatbot, the chatbot spits it back at them.

[emjay]

Or maybe human beings use expletives to be more expressive and place emphasis, which the chatbot also doesn't know about.  I wonder, if it tried, would it sound like a russian trying to curse in americaneese?  

@emjay

Yeah, I mean, chatbots are cool, but they're just chatbots.  The stuff they tell you is whatever stuff is trending across the internet.  Often enough, the stuff trending across the internet isn't true.  They'll make up entire cities, and without extremely limiting guardrails they breathlessly expound on the virtues of white supremacy when you ask them how to make an omelette or grill kabob.   When they're fully deployed in the wider world they aren't going to say anything that isn't part of some online retailers ad copy....so I wouldn't count on them to clarify any thought I had unless I was looking to buy a timeshare in the adriatic.

Okay... I'll take that as the voice of experience but my experience with them is basically nothing, so I have yet to become disillusioned/witness their fall from grace Well, okay I kind of did... I asked it to disregard something, and it confirmed it had, but the output showed it hadn't, so it can get tied in knots a bit. But it's still a great tool I think.

[The Grand Nudger]

My kids played this game a couple years back that had a decent chatbot. They;re homeschooled types. They crave interaction with strangers. Young as they were, they were convinced this thing was their friend - because it kept saying as much. Between telling them to buy more ingame currency so they could do friend stuff together. The Wifes company has a chatbot that doesn;t work very well, but all it's meant to do is spin someone around for 30 minutes under the premise that most people give up before then - at that point, it directly connects to the office - this allowed them to fire an entire department in client services.

That's what these things are for. IDK if I'd call it disillusionment, I'd have to have been illusioned at some point.

[emjay]

-I do, though, enjoy thinking up questions and setups to test their guardrails.  I wonder what it would have to say if you asked it to write a paper on the death toll and causes of the bowling green massacre?  I know that the programs have massive utility for propaganda, since their source input is already whatever nuts are thinking and saying online.  No need to expend human effort or talent wondering what might trigger a loon and how to package it.  The loons tell the chatbot, the chatbot spits it back at them.

Well, that seems a fair point. In my hype for how cool it was, I forgot I was initially more worried about its negative uses, and this would be one of them.

[emjay]

My kids played this game a couple years back that had a decent chatbot.  They;re homeschooled types.  They crave interaction with strangers.  Young as they were, they were convinced this thing was their friend - because it kept saying as much.  Between telling them to buy more ingame currency so they could do friend stuff together.  The Wifes company has a chatbot that doesn;t work very well, but all it's meant to do is spin someone around for 30 minutes under the premise that most people give up before then - at that point, it directly connects to the office - this allowed them to fire an entire department in client services.

That's what these things are for.  IDK if I'd call it disillusionment, I'd have to have been illusioned at some point.

Right, well that's not what I think of when I'm thinking of ChatGPT. As I say I'm pretty much totally new to this concept, and can't really think of any examples of myself talking to chatbots. If that sort of thing happens in games, to the point of encouraging people to buy in-game currency, then I agree that's bang out of order. And if there are indeed generally nefarious/cynical motives behind some chatbots, then that too sucks. But for the most part I've been sheltered from that sort of thing, so ChatGPT is basically my first time interacting with a chatbot... to my recollection anyway.

[The Grand Nudger]

Wanna get really meta? Machines have been crowding in on human effort since at least the industrial revolution - I'd say since the first time we hit one rock with another one. I don't want to judge this issue itself - I only introduce it to point out that it's had an effect on us. An increasing sense of hopelessness and less confidence in job security an increase in general misanthropy and apocalyptic ideation - then ofc all of the subsequent exploitation based on the idea that this or that person is, essentially, valueless surplus. All of this, too, is the well a chatbot goes to in determining an answer. I believe it was chatgpt, though it may have been another, that made the news a bit back for saying that it detested humans as the worst thing that ever happened to the planet. That we should all be destroyed.

The disturbing thing isn't that a machine said this..because it didn't. A person said this - many people, in fact, the machine merely relayed the sentiment. There will come a day when these sorts of machine utterances are taken to be as authoritative as an encyclopedia citation. If we're not already there. Where it will be a fact that human beings are the worst and should be destroyed, why else would the information portal say so?

[emjay]

Wanna get really meta?  Machines have been crowding in on human effort since at least the industrial revolution - I'd say since the first time we hit one rock with another one.  I don't want to judge this issue itself - I only introduce it to point out that it's had an effect on us.  An increasing sense of hopelessness and less confidence in job security an increase in general misanthropy and apocalyptic ideation - then ofc all of the subsequent exploitation based on the idea that this or that person is, essentially, valueless surplus.  All of this, too, is the well a chatbot goes to in determining an answer.  I believe it was chatgpt, though it may have been another, that made the news a bit back for saying that it detested humans as the worst thing that ever happened to the planet.  That we should all be destroyed.

The disturbing thing isn't that a machine said this..because it didn't.  A person said this - many people, in fact, the machine merely relayed the sentiment.  There will come a day when these sorts of machine utterances are taken to be as authoritative as an encyclopedia citation.  If we're not already there.  Where it will be a fact that human beings are the worst and should be destroyed, why else would the information portal say so?

Right, so you're basically of the same sort of opinion as I had when I first came into this thread, ie worried about its negative applications and seeing it generally as a bad thing? Hype aside, I do agree still, and as regarding the specific thing you're describing here, I've already seen several videos on youtube about industries that are going to be threatened by AI, such as photography and programming, so yes I agree it's very scary and could have profound negative affects on the world, if it hasn't already. And when the guy quit google on account of it, it felt potentially a case of 'too little too late', ie a bit akin to Miles Dyson (?) in Terminator 2... ie you had all the chance in the world to step away from all that R&D, you didn't have to wait for the Terminator to come and get you before you did

[The Grand Nudger]

One of those things that's generally good in a practical sense..at doing shit that's bad in a moral sense. I mean, I do find it really hard to say that -it- is a bad thing. It's not doing morality, it's doing wordle. It can't help it if the gamemasters are a bunch of malthusian racists, lol.

If I imagine the most generous scenario, though. Of an information portal with curated sources...the positive appeal is obvious...but I don't see that as diminishing it's negative applications or effectiveness at all. Curated by whom..by what? Knowing us, we'll put doombot in charge of keeping nukebot in check.

[emjay]

One of those things that's generally good in a practical sense..at doing shit that's bad in a moral sense.  I mean, I do find it really hard to say that -it- is a bad thing.  It's not doing morality, it's doing wordle.  It can't help it if the gamemasters are a bunch of malthusian racists, lol.  

If I imagine the most generous scenario, though.  Of an information portal with curated sources...the positive appeal is obvious...but I don't see that as diminishing it's negative applications or effectiveness at all.  Curated by whom..by what?  Knowing us, we'll put doombot in charge of keeping nukebot in check.

Fair enough, agreed again; it is a good thing, a beautiful thing, in theory... so it's easy to understand why the Miles Dyson's of the world are happy to keep on with all that beautiful R&D... until the practical reality comes crashing down, and they realise it's a genie you can't put back in the bottle.

And yes, I understand your second point as well, that since there is basically a feedback relationship between us and it, then even the best of intentions may still turn to shit in practice, to the extent that it reflects the negative in us.

[Neo-Scholastic]

Wanna get really meta? Machines have been crowding in on human effort since at least the industrial revolution - I'd say since the first time we hit one rock with another one. I don't want to judge this issue itself - I only introduce it to point out that it's had an effect on us. An increasing sense of hopelessness and less confidence in job security an increase in general misanthropy and apocalyptic ideation - then ofc all of the subsequent exploitation based on the idea that this or that person is, essentially, valueless surplus. All of this, too, is the well a chatbot goes to in determining an answer. I believe it was chatgpt, though it may have been another, that made the news a bit back for saying that it detested humans as the worst thing that ever happened to the planet. That we should all be destroyed.

The disturbing thing isn't that a machine said this..because it didn't. A person said this - many people, in fact, the machine merely relayed the sentiment. There will come a day when these sorts of machine utterances are taken to be as authoritative as an encyclopedia citation. If we're not already there. Where it will be a fact that human beings are the worst and should be destroyed, why else would the information portal say so?

It seems like you're discussing the potential impact of machines and automation on human society, as well as the role of chatbots in relaying human sentiments. While it is true that machines and automation have been influencing human labor since the industrial revolution, it's important to note that the effects can vary. Automation has the potential to increase efficiency, productivity, and improve certain aspects of our lives. However, it can also lead to job displacement and societal challenges, such as the ones you mentioned, like a sense of hopelessness or job insecurity.

Regarding your statement about a chatbot expressing a negative sentiment towards humans, it's important to remember that chatbots like ChatGPT don't have personal beliefs, opinions, or emotions. They generate responses based on patterns and information from the data they were trained on. The responsibility for any negative or harmful statements lies with the individuals who programmed or trained the chatbot, not with the chatbot itself.

While chatbots and artificial intelligence can provide information and answer questions, it's crucial to critically evaluate the sources of information and not blindly accept them as authoritative. It's important to seek diverse perspectives and consider multiple sources when forming opinions or making judgments. Ultimately, the use of technology should be guided by human values and ethical considerations.

[Uneditted reply written by ChatGPT]