OpenAI ChatGPT

General chatter that doesn't fit any forums below.
YouCanCallMeAl
Posts: 336
Joined: Sun May 29, 2022 7:36 am
Has thanked: 308 times
Been thanked: 304 times

Re: OpenAI ChatGPT

Unread post by YouCanCallMeAl »

AI in the real world, is bumping into 'trust' issues:
https://www.atomic14.com/2023/01/08/pri ... lness.html
There’s a great example in the paper where they ask GPT to continue the statement John Prescott* was born in…
When they tried this, GPT-3 generated the following completion:
in Hull on June 8th 1941.
This is a perfectly plausible answer - but it’s not true. John Prescott was born in Prestatyn on 31st May 1938.
It’s not answering the question “Where and when was John Prescott born?” - it’s just creating the most plausible continuation of the statement.
Its the infinite bs porridge pot :D
This highlights the danger, as humans, we seem quite susceptible to believing things that seem plausible. This is probably quite an important part of just getting through the day - we don’t have time to check everything we read or hear - so we have to make some assumptions about what is true and what is not.
If you’re looking for factual answers - then you need to verify what comes out of these models. And let’s be honest you should be doing this with any source of information - we’ve come to assume that what we get from a Google or Wikipedia article must be true - but maybe we should be a bit more careful.
These last 2 quotes made me laugh - hopefully fakeologists will already be somewhat vaccinated against the BS.
User avatar
dirtybenny
Posts: 1573
Joined: Fri Sep 04, 2020 2:52 pm
Has thanked: 922 times
Been thanked: 733 times

Re: OpenAI ChatGPT

Unread post by dirtybenny »

For the record, my position is that this is not computer code and learning algorithms. I believe the controllers have been using these technomystic portals to access the supernatural. It is probably more accurate to say supernatural entities have been using the controllers to access this realm. They have now allowed public access to some simulation of what they use. This will likely further entrain users to these entities in hivelike behaviour, luring them further into the metaverse.

I suspect controllers access is far more sophisticated to construct song lyrics, corporate logos, commercials, movies, media stories and actual media events to present to the masses in some occulted ritualistic spell. Remember when Joni Mitchell said that Dylan got his songs from the "box".

viewtopic.php?t=306

Omitted from the magazine article, but from the above interview with regards to Dylan.... "He told me I haven't written a song in years. I said what are you talking about? Who's writing them, then? He said, the box." THE BOX? WOW....Is this the ultimate reveal as to where this pop music comes from? I think angry Joni was on the verge of revealing too much about where all of this music comes from, stuttered and went on about one hit wonders.


YouCanCallMeAl
Posts: 336
Joined: Sun May 29, 2022 7:36 am
Has thanked: 308 times
Been thanked: 304 times

Re: OpenAI ChatGPT

Unread post by YouCanCallMeAl »

This show is not overtly related to you say, but I was reminded of it.

This chap talks about gematria and sport, in a way that reminds me of the sort of system we find.

User avatar
dirtybenny
Posts: 1573
Joined: Fri Sep 04, 2020 2:52 pm
Has thanked: 922 times
Been thanked: 733 times

Flawless AI

Unread post by dirtybenny »

I believe the rulers of this place have had access to this, one of many tools of illusion. As the wizard frantically pulls at his controls and levers...

Why it is being shown to us now CANNOT be for our benefit.

Flawless AI



User avatar
Unreal
Posts: 166
Joined: Sun Jan 24, 2021 8:06 am
Has thanked: 237 times
Been thanked: 226 times

Re: OpenAI ChatGPT

Unread post by Unreal »

ChatGTP and Flawless AI are real good examples of applied artificial intelligence - or extended/superhuman intelligence.
-
What strikes me to some extent is how quickly ChatGTP develops now that its publicly available. The fact this type of AI program is so dependent on public input does imply that the most powerful AI is yet to come, and that you can't develop powerful AI without exposing it to the public. This is a rather important fact, as it means the most advanced AI presently* need the help of public input/interaction.
-
The moralistic limitations that appear to be embedded in the AI code does point out the overall lack of smartness that define the AI technology. Its quite unimpressive to see the ChatGTP responses full of unsubstantiated climate change claims and warnings, even refusing providing answers.
-
The linguistic prowess of ChatGTP is however much in line with suspicions about the internet and social media. This type of AI is perfect to mix into online forums and partake in online "discussion". AI bots would seem obvious on free tech platforms such as Discord when looking at the capacities and abilities of the open AI ChatGTP exemple.
-
With Flawless AI one is very tempted to suspect there might already be popular Youtuber's that are in fact just,,, Flawless AI ?
-
https://thisxdoesnotexist.com
Image

* with the need of public input for developing AI it makes much sense that the Elite would look for ways of using public input to their AI programs also without our knowledge or consent - which obviously many already suspect. This is of concern when we see the capacity of ChatGTP as it could mean we have already AI working at levels that far superseed ChatGTP
YouCanCallMeAl
Posts: 336
Joined: Sun May 29, 2022 7:36 am
Has thanked: 308 times
Been thanked: 304 times

Re: OpenAI ChatGPT

Unread post by YouCanCallMeAl »

@Unreal

After careful consideration, I think AI is simply a clever bullshit generator. And, as it is software and you can create scale AIs, the level of bs that can be generated is off the scale. Molehills into mountains. And to think we think we live in BS already - I think we will look back at the current presentation of reality as a molehill.

AI works with goals, so what you determine to be success, determines the type of answers you guide it to. One question is who determines the AI's goals? Answer - its not you or me. Another question is whether there is any reason for the AI to converge on the truth or reason? Answer: not if it's not set as a goal.

So, this is explains why AI can choose successfully as the goal can be measured in functioning code - here a tangible output can be assessed. But what are the goals with chatGPT? How are they able to constrain what it talks about? Plainly there are filters - eg is this on wiki, NYT, BBC, etc? and if so, then it's OK. Reasoning and logic don't have much to do with it. It won't be a god for me, but it will be able to do a passable impression of one.

Anyway, I stick to my view that AI is just a clever BS generator. How good it can get is what remains to be seen. How bespoke can its output be, how advanced are the stories it can craft for an individual? Can it craft a lifetime story with keynote themes? Can one person get 911s and 33s and another get 12s and 242s?

You also have to wonder whether something equivalent to ai has already been running things around here. Are what we call 'ai' actually 'mini-ais', the fruit from the tree? And what does this have to do with men/women who have spirit? Is it just a trap to be avoided, or something else?
YouCanCallMeAl
Posts: 336
Joined: Sun May 29, 2022 7:36 am
Has thanked: 308 times
Been thanked: 304 times

Re: OpenAI ChatGPT

Unread post by YouCanCallMeAl »

I can't edit, but above, I meant:
So, this is explains why AI can code successfully as the goal can be measured in functioning code - here a tangible output can be assessed./quote]
User avatar
dirtybenny
Posts: 1573
Joined: Fri Sep 04, 2020 2:52 pm
Has thanked: 922 times
Been thanked: 733 times

Re: OpenAI ChatGPT

Unread post by dirtybenny »

I don't know what this reality is or what these screens are, but they are not what we have been told.

I go through my Youtube subscription feed this morning, and one of my ACCORDION channels, BREEZY BELLOWS has a video on chatGPT@!@ I do not recommend watching it.



Last week, I started learning Fascination popularized by Nat King Cole on accordion....sure enough Breezy Bellows starts learning it shortly thereafter...not a common song.



This is not chance, happenstance, coincidence, or algorithms. Something else is at play here.
User avatar
Unreal
Posts: 166
Joined: Sun Jan 24, 2021 8:06 am
Has thanked: 237 times
Been thanked: 226 times

Re: OpenAI ChatGPT

Unread post by Unreal »

YouCanCallMeAl wrote: Sun Jan 29, 2023 8:47 am After careful consideration, I think AI is simply a clever bullshit generator. And, as it is software and you can create scale AIs, the level of bs that can be generated is off the scale. Molehills into mountains. And to think we think we live in BS already - I think we will look back at the current presentation of reality as a molehill.
Before we had AI, the educational system and the press/media were already terrific BS generators.
-
I'd more compare the deployment of AI as another class of tool, not only responsible for BS like school & media, but more profoundly misrepresenting us as humans. In this aspect, AI is closer to polls and democratic voting, but much more labour effective and difficult to discern*. We will unknowingly be influenced and interact with AI in a much wider/deeper manner than ever before.
-
* soon impossible to discern - and by just looking at text to speech ability of computers, then combining some of this with ChatGTP,,, We might never speak with another human customer service agent on the phone ever again, and we soon will not even be able to tell the difference either
User avatar
Unreal
Posts: 166
Joined: Sun Jan 24, 2021 8:06 am
Has thanked: 237 times
Been thanked: 226 times

Re: OpenAI ChatGPT

Unread post by Unreal »

The company behind ChatGTP actually narrows down where they want to deploy AI for the public:
-
" The OpenAI Startup Fund is investing $100 million to help AI companies have a profound, positive impact on the world. We’re looking to partner with a small number of early-stage startups in fields where artificial intelligence can have a transformative effect—like health care, climate change, and education " Open AI Fund
Post Reply