Cassiopaea and ChatGPT

Laura said:
Told you it was dumb.
Click to expand...
It can be helpful, it can be literally dumb.
Sometimes you can see how it's going to get into some dead end. But sometimes it's even better than I expected. I think it's a bit overrated at the moment, but that also shows how used to conveniences we've become (after all, in the past, such an invention would have been the eighth wonder of the world).
 
After I read Laura's article about AI and Shakespeare I had to try. I have used Flux Image Generator and this what I got. I know, some things are weird:-D.

1. AI
2. Albino Corn Snake (I asked just for one head :scared:)
3. Fantasy creature, alien, big claws but I didn't ask for two tails :shock:)
4. Weird stuff (that's right, I just wrote "weird stuff")
5. Klaus Schwab (every time I asked for the Schwab it took him so much time to generate this guy)
6. Victoria Nuland (hah! Such a nice lady:cool2:)
7. Monica Bellucci
8. Eiffel Tower (again, I asked for one, got two)
9. Conspiracy theory
10. Cigarette
11. George Soros

I'm gonna play some more but like someone said it's addictive but could be great tool.
 

Attachments

  • Blue Green Natural Travel Photo Collage.png
    Blue Green Natural Travel Photo Collage.png
    3.8 MB · Views: 30
After all, it is a world in duality, but chwobb and shwaros seem the same families, hehe. As for the cigarette is it a kind of firecracker? It’s still stupid and stupid... An animal is smarter
In any case it is certainly not La Bellucci! ;-D
 
Ina said:
Knight-Jadczyk discusses how mysticism and astrology continue to influence Christian thought and practice.
Click to expand...
Laura didn’t discuss the Gospel of John or astrology per se that much in FPTM if I remember correctly. The output seems a bit ‘boiler plate’. It seems one has to drill down from general outline to specifics to get specific info. AI seems programmed with a lot of canned common knowledge, assumptions, and generally accepted facts and rules. it uses left-brained thinking. It has one giant statistical model that wants to fit every situation. The problem with such models is that they will be wrong some percentage of the time, and no one can know exactly when the model will fail any individual case. The user needs to get the AI to drill down and focus on the specifics of the inquiry, it has to be coaxed into seeing the uniqueness. It sees individual things… or people as ‘instances of a class’, AKA an ‘object’, where the object is an avatar running in the ‘the program’ real-time. A class is a canned definition of something… a human or whatever, stored in memory. However huge these class definitions may be, I doubt they can define anything remotely complex completely or even accurately, at least not all the time.
 
I'm thinking along the line of AI proliferation and democratization... One of possible consequences should be the cost lowering of printed old books and / or the disappearance of libraries, first the small ones then the big ones, unless they'll be turned into museums of sorts. Perhaps we should start a quiet strategy for a Save the Books movement. Incognito. I wonder how many books can we save?
 
You must log in or register to reply here.

Trending content

Back
Top Bottom