• Welcome! The TrekBBS is the number one place to chat about Star Trek with like-minded fans.
    If you are not already a member then please register an account and join in the discussion!

AI / ChatGPT Discussion

Gingerbread Demon

Yelling at the Vorlons
Premium Member
So I tried Chat GPT with three Star Trek questions.

Not a bad little AI, very diplomatic little critter it is..

I have to say these in hindsight were kind of what I thought the AI might say, I really did want to be surprised but still thought the responses were delightful.

STARTREKQUESTION01.jpg STARTREKQUESTION02.jpg STARTREKQUESTION03.jpg
 
Last edited by a moderator:
See, I think the whole topic around AI generally or ChatGPT in particular is an area ripe for discussion. There are many possible avenues for such a discussion to take, such as impacts on artists now that AI is producing art, impact on our jobs and society as a whole now that AI is progressing to a point where they can do the jobs many of us do, moral/ethical implications of AI, etc. Unfortunately, I think you have limited the potential of this thread a bit by such a narrow focus; it doesn't ask any questions, and it's basically just "look what the AI said about Star Trek technology"... so it doesn't really set things up to invite much response. (And Star Trek topics don't really belong in this forum anyway.)

So, I'm proposing that we broaden the scope of this thread to cover AI / ChatGPT topics as a whole. I will rename the thread accordingly (and will move your original thread title into the body of your post so that the context of your post is not lost). Hopefully this is OK with you.

Recent advances in AI are set to significantly change things in our society over the next little while, and I'm not convinced we are prepared for it.

Let's discuss.
 
See, I think the whole topic around AI generally or ChatGPT in particular is an area ripe for discussion. There are many possible avenues for such a discussion to take, such as impacts on artists now that AI is producing art, impact on our jobs and society as a whole now that AI is progressing to a point where they can do the jobs many of us do, moral/ethical implications of AI, etc. Unfortunately, I think you have limited the potential of this thread a bit by such a narrow focus; it doesn't ask any questions, and it's basically just "look what the AI said about Star Trek technology"... so it doesn't really set things up to invite much response. (And Star Trek topics don't really belong in this forum anyway.)

So, I'm proposing that we broaden the scope of this thread to cover AI / ChatGPT topics as a whole. I will rename the thread accordingly (and will move your original thread title into the body of your post so that the context of your post is not lost). Hopefully this is OK with you.

Recent advances in AI are set to significantly change things in our society over the next little while, and I'm not convinced we are prepared for it.

Let's discuss.


Oh no totally agree.

There is a hell of a lot here to discuss, I just chose three examples, so feel free to have at it. Interesting you do get the impression it's thinking because that first question was almost 20 seconds before I got a reply, you can also reset it and have it think up another response if you do not like what it says the first time. But it's a very limited experience at the moment but it's clever I will give them that.

People have been using it to set up quizzes by saying "make me a quiz about x" and such and it generates both the questions and answers, again that's very limited but it's neat for want of a better word.

These things can and will get more advanced given time and resources. It's inevitable and unavoidable. It will affect how we do things in the world as AIs become a bigger part of it, and more ubiquitous in nature, to the point where we might not know if we are interacting with an AI or not, but that is a long, long way away at this point in time I think.
 
I'm not familiar with all the details, but I've seen an issue mentioned when it comes to AI-"generated" art, that some of the AI engines, or whatever you call them, tend to steal parts of existing artwork instead of actually fully "generating" something new.

Kor
 
Yeah, I think part of the problem lies in how they've been trained, ie feeding lots of random art to them from different sources. Of course, an AI system isn't going to know right from wrong and will just gather, unless you tell it not to. Using DALL-E as an example, it seems to be it will search Google for anything that matches the keywords that the user has typed in and tries to do its best. No idea if that stays in the system for followups though. Sites like DeviantArt are now planning on sueing these AI developers. Websites can be defined to prevent searching though, and it seems to me a possible solution would be to have some sort of definable handshake between the AI and the search platform, where a website would have to have a line in its code that disallows an AI from searching through it, while the search engine would have a parameter which the AI would have to call out as a way to identify itself, similar to how a browser identifies itself to search engines.
 
  • Like
Reactions: Kor
It's a great way to address the other contentious issue of "eliminating the idea of homework", since kids doing the work in class prevents them from going to the site to do the research work from (how many sites) for them.

It's still a good time saver for the right venue and, indeed, is more impressive than various tests taken a decade ago that took the received criteria and pulled back paragraphs of text based on said criteria matches, blocking others where needed if various matches were "counterintuitive", throwing out various matches and adding 1 to a counter and returning "inconclusive" if that "AI" didn't know how to categorize because it has no "I". It's just a (n impressive) bunch of algorithms ran under a large series of conditional statements, looking for metadata related to content to pull back. If, in the middle of such a document, there was any real intelligence, it'd do something more than parrot back retrieved documents from a pattern-based/heuristic scan. Real AI wouldn't know what to do with a random variable, stochastically- yet unsarcastically- speaking. Or else, in the middle of the encyclopedic and sourced references about a fictional technology, it'd throw in tips on how to feed elephants to your pet hamster mating and how to clean up the mess of microchips it excretes while playing Fizzbunt, which for Funkytown reasons I didn't use as pragmatic example... :devil:
 
I'm not familiar with all the details, but I've seen an issue mentioned when it comes to AI-"generated" art, that some of the AI engines, or whatever you call them, tend to steal parts of existing artwork instead of actually fully "generating" something new.

Kor

They're definitely influenced by the bit patterns; how they take then amalgamate and randomly rearrange or determine a new output is still impressive. If nothing else, human models won't be needed anymore since the computer can generate a composite from numerous existing static images, or even look at samples of catalogued DNA and come up with an amalgamation. Isn't that how they managed to figure out what ancient pharaohs and others looked like? --> https://www.livescience.com/ancient-egyptian-mummies-faces-reconstructed-dna And how the scientists could figure out what pigments were originally painted on statues via a sample or scan -- https://www.npr.org/2022/07/12/1109...es-werent-white-now-you-can-see-them-in-color ... now that's brill. :techman: Just wait for the polygonal count to improve so they won't look like video game characters from 2015...
 
Yeah, I think part of the problem lies in how they've been trained, ie feeding lots of random art to them from different sources. Of course, an AI system isn't going to know right from wrong and will just gather, unless you tell it not to. Using DALL-E as an example, it seems to be it will search Google for anything that matches the keywords that the user has typed in and tries to do its best. No idea if that stays in the system for followups though. Sites like DeviantArt are now planning on sueing these AI developers. Websites can be defined to prevent searching though, and it seems to me a possible solution would be to have some sort of definable handshake between the AI and the search platform, where a website would have to have a line in its code that disallows an AI from searching through it, while the search engine would have a parameter which the AI would have to call out as a way to identify itself, similar to how a browser identifies itself to search engines.

If one of those sites was Getty they will most definitely be sued for a lot.
 
I'm not sure if they are, but maybe. I could definitely see them being one of them, given that many images appear via search engines when they shouldn't, if they're behind a paywall of some sort. I think a rethink on image searches need to be done. The only images that should would be images that have been marked as 'open' or 'free' licenses by their authors. Even then, you run into the issue of art that is then modified by an AI claiming to be an original work.
 
There's already an AI thread in Sci-Tech.

Ah, I wasn't aware. I don't typically visit that forum.

I'm of two minds of this. On one hand, we probably don't need two threads covering the same topic on the open side of the board. And I have moved sports-themed threads to the sports forum in the past. OTOH, I've heard the argument made that for large or important topics, it's not completely out of line to discuss the topic in more than one forum, in order to cater to people that may hang around in one forum and not the other.

What do you think, denizens of Misc? Are there enough of you here, who don't typically visit Science and Technology, who are interested enough in this thread to make it viable as its own thing? Presumably we wouldn't get too far into the tech side of the topic here, so there might be enough differentiation to warrant it staying. OTOH, if there's not enough local interest here, or if most people here also visit Science and Technology, then it makes more sense to just use the existing tread and close this one. (I can't merge this one with a specific thread in another forum.)

I found the link to the other thread if any of you would like to take a look:
https://www.trekbbs.com/threads/the-artificial-intelligence-thread.303368/

"shrug" @Avro Arrow moved this here

Didn't really move anything, just renamed a thread that already existed here.
 
my question: If time had 3 dimensions the now the future the past dimensions could AI or humans manipulate these dimensions for our own goals or purposes?

AI reply: If time were to have three dimensions, it is likely that our understanding and manipulation of the physical world would be quite different from what it is today. It is possible that with a better understanding of these additional dimensions, we may be able to manipulate them for our own goals or purposes. However, it's important to note that these are currently just theoretical concepts in physics and we don't have any experimental evidence of the existence of extra dimensions of time, so it's impossible to say for certain what the implications of such a discovery would be.

It's also worth noting that even in the current understanding of time as a single dimension, AI and humans have been able to develop technologies that allow us to measure, record, and manipulate time to a certain extent. For example, we have atomic clocks that can measure time with incredible precision, and we can control the speed of time using special Relativity.

regenerated the response again..

In physics, time is typically considered to be a single dimension and it is not possible to manipulate time in the sense of changing the past or the future. However, some theories in physics, such as theories of relativity, suggest that the flow of time may be affected by the presence of matter and energy. For example, time appears to move slower in stronger gravitational fields.

As for AI or humans, it is currently purely fictional idea that we can manipulate time dimension in any way, as of today we don't have any technology to support such manipulation and there is no scientific evidence supporting the existence of extra dimensions of time.
 
Last edited:
my next question: are there multiple alternate quantum dimensions of reality that we don't know about?

AI reply: In quantum mechanics, the concept of parallel universes or alternate dimensions is often discussed in the context of the "many-worlds" interpretation of quantum mechanics. According to this interpretation, every time a quantum event occurs that has multiple possible outcomes, all outcomes occur simultaneously in different universes or dimensions.

This idea is still a subject of debate among scientists and philosophers, and there is currently no experimental evidence to support the existence of multiple parallel universes or alternate dimensions. While the many-worlds interpretation is one possible explanation for the strange behavior of quantum particles, it is not the only explanation and it is not universally accepted among scientists.

It's worth noting that there are some theories of physics, such as string theory and M-theory, that propose the existence of additional dimensions beyond the four dimensions of space-time that we observe in our everyday life. However, these dimensions are not considered to be alternate realities as we know them, but rather additional dimensions of space that are curled up or hidden from our perception.

further added a new question: what is a quantum event?

AI response: A quantum event is an occurrence that takes place at the level of atomic and subatomic particles, which are governed by the laws of quantum mechanics. In classical mechanics, the behavior of particles can be predicted with a high degree of accuracy, but in quantum mechanics, the behavior of particles is described probabilistically, meaning that it is only possible to predict the likelihood of certain outcomes.

A quantum event can refer to various physical processes, such as the decay of a radioactive atom, the absorption or emission of a photon by an atom, or the entanglement of two particles. These processes are described by quantum mechanics laws, which are different from classical mechanics laws, and that's why they are called quantum events.

One of the most famous and fascinating quantum events is quantum superposition, where a quantum object can exist in multiple states or locations at the same time, until it is measured, then it collapses into one state.
 
one more --- I think I am getting the hang of what AI chat can do---

Me: Make metaphorical verses about silliness in a quantum elevator going to heaven.

  1. "In a quantum elevator, silly thoughts do rise, To the heavens above, where logic dies. But the journey to enlightenment, is not one of mirth, For the purest of mind, is the path of rebirth."

  2. "With a press of a button, in a quantum lift, Silly notions take flight, in a cosmic rift. But the journey to heaven, is not one of glee, For the true path to salvation, is humility."

  3. "Quantum elevator, taking us high, Silly thoughts along for the ride, to the sky. But the path to enlightenment, is not one of fun, For the way to salvation, is to be done."

  4. "Up and up we go, in a quantum lift, Silly notions adrift, in a cosmic drift. But the journey to heaven, is not one of jest, For the way to inner peace, is to be at rest."

  5. "In a quantum elevator, to the heavens we soar, Silly thoughts galore, forevermore. But the path to enlightenment, is not one to mock, For the road to inner peace, is one to unlock."
Note that the use of the word "heaven" here is metaphorical, and it's not representative of any specific religious concept or belief. It's used to convey the idea of reaching a high level of understanding or a state of inner peace.


this is fun ---the AI had a disclaimer note too-- just like that it took less than 10 seconds to pump this out OMG--- wow

Me saying to the AI: Make metaphorical verses about the timeless realities that exist between the M-branes and mindless thoughts of beingness.

"Between the M-branes, timeless realities reside,
Mindless thoughts aside, in a cosmic divide.
Eternity's door, forever ajar,
A realm of beingness, where time is afar."

"Beyond the realms, of the M-branes,
Timeless truths, in a cosmic chain.
Mindless thoughts, a fleeting breeze,
Eternity's existence, where mind finds ease."

"M-branes divide, timeless realities unite,
Mindless thoughts take flight, in the cosmic light.
Eternity's realm, a place of stillness,
Beingness, a state of timelessness."

"In the cosmic expanse, between M-branes,
Timeless realities, forever sustains.
Mindless thoughts, a mere distraction,
Eternity's embrace, a state of satisfaction."

"M-branes divide, timeless realities abide,
Mindless thoughts, a cosmic tide,
Eternity's realm, a state of grace,
Beingness, an eternal embrace."

It's worth mentioning that the concept of M-branes is a theory in theoretical physics and it's not yet established as a reality, but it's used as a metaphor here to convey the idea of a different reality or different dimension existing beyond the one we perceive. Also, beingness here is used as a state of being, existing in a timeless and eternal way.

AI is like yeah --- I can't stop asking for new AI verse --- :)
 
Last edited:
I guess some people have tried to make it trip up by asking it the obvious questions.
"Are you aware of yourself?"
"Do you know what you are" that kind of thing. I wonder how it might respond
 
@think , just a gentle reminder that we shouldn’t be posting more than twice in a row. Thank you.

I did not realize it was that many times,. I should of edited my first post adding it with the other few posts,. sorry about that --- the verse poems the AI was making was so cool for me .. -- sorry,..
 
I've tried conversing with some newer AI chat bots. They seem to quickly forget stuff that they said just a couple sentences before, and therefore contradict themselves a lot. I think it's going to be a while before they can be really convincing.

...

What do you think, denizens of Misc? Are there enough of you here, who don't typically visit Science and Technology, who are interested enough in this thread to make it viable as its own thing? Presumably we wouldn't get too far into the tech side of the topic here, so there might be enough differentiation to warrant it staying.
...
I don't visit the sci & tech forum much because I feel like I don't have a solid enough grasp on the hard-science-y stuff for the conversations there. In this forum, we could get more into cultural and social implications, and so on.

Kor
 
If you are not already a member then please register an account and join in the discussion!

Sign up / Register


Back
Top