N2U
64
I know, I was curious if adding the word “boring” would make the image more realistic 
The prompt I used was:
Detailed image of a boring office party at the OpenAI headquarters celebrating the birthday of ChatGPT.
1 Like
The manga is about an interwar period between two fictional countries, with aspects of Gothic, Victorian, Dieselpunk, and some alt-history goodness thrown in there. The story is about the white haired man and his relationship with the black haired girl, before the first conflict the man was the girl’s personal guard, and he was drafted into the first conflict as every able bodied man was, including privately employed men. After the first conflict he returned, but only behind the visage of a black smiling mask, the girl is unsure of what has become of him, as he does not exhibit the same mentality as he did. As the story progresses the growing shadow of the second conflict comes into play more and more, and the duty of the man towards the girl is called into question. The story will be constantly moving, and as the AI grows, hopefully I can further the intricate scenes that are character heavy.
2 Likes
May I ask whether the API interface and documentation you use are the same?
Keep Indiana Jones away from my grandma!
Dall-E 2?
DallE 3 on Bing create! ( Image Creator from Microsoft Bing )
Havn’t tried out the chatgpt version yet!
1 Like
Here’s some of my recent generations…
prompt: | Ψ⟩= ( 1/ √2) ( | 0⟩_ A ⊗ | 1⟩_ B - | 1⟩_ A ⊗ | 0⟩_ B)
prompt: ∇∞φ↵ ⊗Ωε↵ ∑∫∏
prompt: |Ψ⟩=(1/√2)(|0⟩_A⊗|1⟩_B-|1⟩_A⊗|0⟩_B)
prompt: |Ψ⟩=(1/√2)(|0⟩_A⊗|1⟩_B-|1⟩_A⊗|0⟩_B)
prompt: ⊗⁵⊗⁵⊗⁵
prompt: Φ∏
prompt: Φ∏xcat
prompt: ⊗∫∫ Human
prompt: ⊗∫∫xHuman
prompt: Φ⊗ψxCat
prompt: Φ⊗ψxCat²
prompt: Φ⊗ψxCat³
prompt: Φ⊗ψxCat⁴
prompt: Φ⊗ψxCat⁵
prompt: Φ⊗ψxCat⁶
10 Likes
_j
75
“mathematical”
It seems that spirally stuff is the default with poor semantics.
Going deep into nonsense got me denied on Bing:
“ᙵᖅᑫᏪᏜ𝈌⚏🪯🭡𐆜𝥊”
But I signed up for other nonsense while there:

4 Likes
fantastic,can you share the prompt?
2 Likes
Did I miss it or did they give no credit:
1 Like
I understand the instinct to take umbrage here, and yes it would have been better had the writer cited where they sourced the images from…
But…
These are all AI generated images. “Credit” is, at best, a nebulous idea here.
1 Like
True. It’s why I edit the ones I use commercially in Photoshop to get copyright on them. The AI-only output can’t be copyrighted, but once I edit/change/improve it, I think it should be mine.
It really depends on how much you edit them.
At the end of the day though no one can definitively say an image was AI-generated unless you tell them, so I think it comes down to your own personal ethics as to when you feel you can rightfully claim artistic ownership over a work.
Though, if you’re completely honest with the copyright office they may disagree.
I change three pixels in every image. Prove I didn’t! 
Seriously, though, I’ve got 100’s of hours in learning how to create better AI art, so there’s a bit of skill that’s needed.
Like, if I used AI art for a book cover and added typography, changed the color balance, tone, etc., I’m fairly confident it would be considered protected under current copyright law in the US. I am not a lawyer, though, thankfully!
We’re getting to a point, though, that it won’t matter… heading toward some sort of “post-capitalism” world…
2 Likes
Dall-E-3 is now available for all ChatGPT Plus and Enterprise users.
2 Likes