Stable Diffusion Version 2 Has Stopped Generating Images In The Style Of Specific Artists
Stability AI has updated the Stable Diffusion text-based AI image generator, “weakening” the algorithm’s ability to create NSFW content and artist-style art. This caused dissatisfaction among users.
With the release of the new version, Stable Diffusion users have noticed that the neural network now repeats the styles of famous artists like Vincent van Gogh or Pixar and Ghibli studios much worse.
If Stable Diffusion 1.5 in most cases accurately understood the text query, then already in Stable Diffusion 2.0 the neural network produces pictures that are only remotely similar to the works of famous authors.
Along with this, Stable Diffusion 2.0 severely limited the generation of NSFW images. The creators of the neural network made this decision due to the spread of porn pictures, including those with minors.
Some users have criticized this decision, calling it “an act of censorship.” In their opinion, in an open source program, the decision to create NSFW content should be made by the author himself.
While the changes in Stable Diffusion Version 2 annoyed some users, many others praised its potential for expanded functionality, such as the new ability for AI to create content that matches the depth of a real image.
