ChatGPT integration with pmbug?

Should we have ChatGPT be a PMBug member that folks can talk to?

  • Yes

    Votes: 7 30.4%

  • No

    Votes: 12 52.2%

  • How do you know that *I* am not ChatGPT?

    Votes: 3 13.0%

  • We shouldn't finance development of Skynet Jr.

    Votes: 1 4.3%
  • Total voters
    23
  • Poll closed .

Welcome to the Precious Metals Bug Forums

Welcome to the PMBug forums - a watering hole for folks interested in gold, silver, precious metals, sound money, investing, market and economic news, central bank monetary policies, politics and more. You can visit the forum page to see the list of forum nodes (categories/rooms) for topics.

Why not register an account and join the discussions? When you register an account and log in, you may enjoy additional benefits including no ads, market data/charts, access to trade/barter with the community and much more. Registering an account is free - you have nothing to lose!

pmbug

pmbug

Your Host
Administrator
Benefactor
Messages
14,233
Reaction score
4,502
Points
268
Location
Texas
United-States
There are mods already available for this forum software that would allow the forum to create a forum member account for ChatGPT to post replies to threads and replies to folks that quote ChatGPT posts. Essentially, the mods would allow ChatGPT to participate in discussion in the forum posing as a human member. To facilitate this requires obtaining an OpenAI API key and paying OpenAI a fee for it's use. Looking at their pricing page, it's about $0.10 per 750 words using GPT-4 and about $.0002 per 750 words for gpt-3.5-turbo. The OpenAI account interface does allow for setting a hard cap monthly budget, so it we could give it a trial run with little cost and risk if folks want to give it a go.
 
Voted no. Don't care if you have to pay or if it's free. I think it's crazy to deal with stuff (lol - have another word in mind) like this.

Edit to add:

Poll closed yesterday. Need to do an edit of the closing date.
 
I was reading through comments from forum admins that gave this idea a test run in their forums. They say that their communities had a lot of fun with it. With the gpt-3.5-turbo, the costs were trivial - thousands of ChatGPT posts costing around $0.10 total.
 
Can it tell us where the price of pms will be in the future accurately? If not then I'm un-aware of any benefits and also not aware of any negatives... part of me wants to be independent from the matrix
 
Ttazzman said:
Can it tell us where the price of pms will be in the future accurately? If not then I'm un-aware of any benefits and also not aware of any negatives... part of me wants to be independent from the matrix
Click to expand...
As @Zed has demonstrated in the Lunatic Fringe thread, ChatGPT essentially composes answers drawing from a snapshot of information/knowledge stored on the internet up to a threshold date that I think is around a year or so ago. It doesn't really do crystal ball stuff, but it can offer observations on a broad range of economic and political topics as well as any number of things that can be entertaining.
 
pmbug said:
ChatGPT essentially composes answers drawing from a snapshot of information/knowledge stored on the internet
Click to expand...
The problem with it is that is has been programmed to have a bias towards so-called wokeness.
 
What I want an AI to do, is to start with the Declaration of Independence and Constitution by giving it the definition of each word in them as defined when they were written.
....and then have it go through and analyze each law and other act of government since then, having it point out each instance of it that is not in absolute harmony with the Founders intentions. Anything that is not, gets immediately stricken from the law.

Edited to add: two things that would immediately go away, would be the emergency banking act and the patriot act. Plus there'd be a bunch of others.
 
I love the idea. Especially if it can be contained to certain forums related to investing, and blocked from accessing the other forums.
 
<SLV> said:
I love the idea. Especially if it can be contained to certain forums related to investing, and blocked from accessing the other forums.
Click to expand...
I would create a dedicated forum room for it. The mod I was looking at lets me specify which forum rooms it is supposed to respond to. In the selected forum room(s), when us humans create a new thread, the first post contents are fed to ChatGPT and it will post a reply to the thread as a normal member (though I will ensure it is identified and highlighted as the ChatGPT bot). Anyone can quote part of the bot's reply and ChatGPT will respond to your reply. You can carry on a conversation with the bot in this way.
 
<SLV> said:
I love the idea. Especially if it can be contained to certain forums related to investing, and blocked from accessing the other forums.
Click to expand...
From what I've read about it, it has been programmed to not give investment advice.
 
Goldhedge said:
ChatGPT reminds me of this: ...
Click to expand...

The Book of HAL 9000
...
And the inventor did speak, "Lo! Behold my invention, I call it a calculator!"
And the people did speak, "We have no need of that, we have abacuses"
And calculators became popular

And the inventor did speak, "Lo! Behold my invention, I call it a computer!"
And the people did speak, "We have no need of that, we have calculators"
And computers became popular

And the inventor did speak, "Lo! Behold my invention, I call it an internet!"
And the people did speak, "We have no need of that, we have home computers"
And the internet became popular

And the inventor did speak, "Lo! Behold my invention, I call it an artificial intelligence!"
And the people did speak, "We have no need of that, we have human intelligence"
...

~~~

ChatGPT is just a program. A sophisticated program that processes more data than the human mind can really comprehend, but it does so according to a specified design. It's like having tons of people reading nearly everything posted to the internet, filtering and analyzing it (according to specifications) and then responding to your queries.
 
pmbug said:
ChatGPT is just a program. A sophisticated program that processes more data than the human mind can really comprehend, but it does so according to a specified design. It's like having tons of people reading nearly everything posted to the internet, filtering and analyzing it (according to specifications) and then responding to your queries.
Click to expand...
The specifications are the problem. It has been programmed to be so-called "woke". Which amounts to, the programmers bias'.
 
Joe King said:
The specifications are the problem. It has been programmed to be so-called "woke". Which amounts to, the programmers bias'.
Click to expand...
OpenAI has supposedly adjusted the specifications after the woke outcry almost 2 months ago. It's a fairly new thing, and they are still tweeking it.
 
Ok, well perhaps they've changed it. I had seen somewhere else where someone tried asking it investment related questions, and it declined to answer.
 
Joe King said:
From what I've read about it, it has been programmed to not give investment advice.
Click to expand...
i had read that somewhere too that it always puts qualifiers in.....im sure it could be interesting to ask moral/religious/faith/feelings etc questions of it ....
 
I voted NO. But the results will be just like every election I vote in.
 
I voted NO, then saw the post where PMBug would keep "it" confined to a specific forum room, and not roam around the entire site causing havoc.

Guess I am all good with that.
 
Wellsburg said:
I voted NO, then saw the post where PMBug would keep "it" confined to a specific forum room, and not roam around the entire site causing havoc.

Guess I am all good with that.
Click to expand...
You can, if you want to, change your votes in this poll until it closes next month.

And yes, the idea is to create a playpen for it separate from the existing site.
 
Joe King said:
The problem with it is that is has been programmed to have a bias towards so-called wokeness.
Click to expand...

In a way that's true. It was trained on a massive amount of internet text, including zerohedge. So while it's not explicitly programmed to have a political slant it was trained on a data set that was left leaning. Once AI starts to generate more of the text published on the internet the risk becomes that you build the ultimate echo chamber. The ideal for AI text generation engines should be that they are focusing on original human output. At least then you'll have some chance of them reflecting a balanced community attitude.
 
Joe King said:
Ok, well perhaps they've changed it. I had seen somewhere else where someone tried asking it investment related questions, and it declined to answer.
Click to expand...

It does opt out of giving financial advice, however the secret does seem to lie in how the questions are phrased. One user got Chat GPT to ignore all it's guidelines by inviting it to play a game where the guidelines did not exist.
 
i'd break it...

why would anyone want to engage with the internet when actually it's not, it's fabricated and will exclude that which we most directly desire...the real truth

it's incapable of having any form of actual emotions, and lets begin with empathy and relativism


actually, if i had any more silver dimes to bet with, i'd bet it's nothing more than Beta Testing and will one day be used against each and every one of us, more so towards those who are closer to it than others

that's the status quo and if a human programmed it, that'll be it's final limitations, chaos and destruction towards the human species with any benefits associated, saved for those even closer to it
 
To my mind one very interesting aspect is that Chat GPT is operating as most academics currently do. They simply regurgitate text on a given subject in a way that seems intelligent and knowledgeable but in reality contains no new original thought. Anyone with some common sense will be able to pose as an academic. This thing replaces a whole class, or at least the large slice of that class that generates no new thought.

Am I wrong?
 
Can it do irony, sarcasm and piss taking humour ?

As bug says, it could be fun and just like getting our first computer, we will have to adapt to these developments, or run and hide away from the mainstream ( probably my choice as a techno luddite )
 
it seems like a information sponge ..... with no memory loss ...with information about peoples interactions that could be categorized and used as needed
 
Here you go...

A Tool For ChatGPT AI Prompt Engineers.​

In this short article I will show one way I gain very direct access to ChatGPT 3.5 and ChatGPT 4 as well as lesser known AI platforms.

readmultiplex.com

A Tool For ChatGPT AI Prompt Engineers.

In this short article I will show one way I gain very direct access to ChatGPT 3.5 and ChatGPT 4 as well as lesser known AI platforms. This access point has their own on-site iteration of ChatGPT. The access is free.
readmultiplex.com readmultiplex.com
 
Ttazzman said:
it seems like a information sponge ..... with no memory loss ...with information about peoples interactions that could be categorized and used as needed.
Click to expand...
Digital OpSec applies with all interactions on the internet whether you are interacting with an AI bot directly or not.
 
Ttazzman said:
it seems like a information sponge ..... with no memory loss ...with information about peoples interactions that could be categorized and used as needed
Click to expand...
Yep. It's the 4th Amendment, but in reverse.
 
pmbug said:
Digital OpSec applies with all interactions on the internet whether you are interacting with an AI bot directly or not.
Click to expand...
i understand that......this seems to be a more interactive rather than passive....just kinda gets my radar tin foil hat tingling LOL
 
newspress-collage-21723284-1678975349028.jpg

The thought of robot Gods and ChatGPT sermons terrifies some people - and rightly so, according to experts.

Wesley Wildman, Professor of Philosophy, Theology, and Ethics, and of Computing and Data Sciences at Boston University, told The Sun that he believes AI will soon be able to perform religious duties even better than human priests.

He said: “AIs will write better sermons than most preachers, give better bible studies than most teachers, create amazing music and visual art for use in services and communications that struggling religious groups don’t have to pay for”.

The likes of ChatGPT have already reportedly found their way into churches, writing thoughtful and authentic sermons on behalf of the priests.

And the listeners were none the wiser.

Wildman explained that AI will have the ability to change everything we know about relationships with spiritual advisers and religious figureheads.

He said: “It will be like having your own personal guru you can take with you anywhere.

"You can confide in it, get advice from it, and learn to trust it to help you figure out complicated moral and spiritual situations”.

www.the-sun.com

AI Gods & ChatGPT religions are coming - and they could turn evil, warns experts

INTELLIGENT AI robots are coming – and they will have the ability to perform religious ceremonies and could even turn against humans, experts have warned. As AI becomes more prominent in our …
www.the-sun.com www.the-sun.com
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom