Found it first here - https://mastodon.social/@BonehouseWasps/111692479718694120
Not sure if this is the right community to discuss here in Lemmy?
this post was submitted on 03 Jan 2024
607 points (93.5% liked)
Technology
59311 readers
6258 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
They need AI to detect the AI images
This is why Stable diffusion images are digitally watermarked. To avoid training it on itself.
Thats not how it works at all
https://codoraven.com/blog/ai/stable-diffusion-the-invisible-watermark-in-generated-images/
thats almost a year old. and their github 2 years. also i did train multiple loras with only ai generated images with sd1.5 and XL without any issue
It's not that you can't train using generated images, but the watermarking is there to allow you to filter training data for them.
Still its neither implemented in a1111,fooocus and comfyui. I acknowledge, it exists and certainly works, but was not adapted by any tool i know of
I have the option to download the images I created with Stable Diffusion with or without a watermark overlay.
Not a visible watermark
https://codoraven.com/blog/ai/stable-diffusion-the-invisible-watermark-in-generated-images
I imagine there are lots of reasons.