this post was submitted on 03 Jul 2025
843 points (95.2% liked)
196
18014 readers
649 users here now
Be sure to follow the rule before you head out.
Rule: You must post before you leave.
Other rules
Behavior rules:
- No bigotry (transphobia, racism, etc…)
- No genocide denial
- No support for authoritarian behaviour (incl. Tankies)
- No namecalling
- Accounts from lemmygrad.ml, threads.net, or hexbear.net are held to higher standards
- Other things seen as cleary bad
Posting rules:
- No AI generated content (DALL-E etc…)
- No advertisements
- No gore / violence
- Mutual aid posts are not allowed
NSFW: NSFW content is permitted but it must be tagged and have content warnings. Anything that doesn't adhere to this will be removed. Content warnings should be added like: [penis], [explicit description of sex]. Non-sexualized breasts of any gender are not considered inappropriate and therefore do not need to be blurred/tagged.
If you have any questions, feel free to contact us on our matrix channel or email.
Other 196's:
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Well those things aren’t generative AI so there isn’t much of an issue with them
What about 'edge enhancing' NNs like NNEDI3? Or GANs that absolutely 'paint in' inferred details from their training? How big is the model before it becomes 'generative?'
What about a deinterlacer network that's been trained on other interlaced footage?
My point is there is an infinitely fine gradient through time between good old MS paint/bilinear upscaling and ChatGPT (or locally runnable txt2img diffusion models). Even now, there's an array of modern ML-based 'editors' that are questionably generative most probably don't know are working in the background.
Id say if there is training beforehand, then its “generative AI”
Not a great metric either, as models with simpler output (like text embedding models, which output a single number representing 'similarity', or machine vision models to recognize objects) are extensively trained.
Another example is NNEDI3, very primitive edge enhancement. Or Languagetool's tiny 'word confusion' model: https://forum.languagetool.org/t/neural-network-rules/2225