People betting on all “lossy compression” being equivalent will lose. It’s a lame argument. Can you encode images to be basically 1:1 with embeddings? Absolutel

People betting on all “lossy compression” being equivalent will lose. It’s a lame argument. Can you encode images to be basically 1:1 with embeddings? Absolutely. Is that what these embeddings are doing? No. If artists try to argue that all copies are equivalent, but are unable to recreate their works from the embeddings, their argument will fall flat. This argument also only applies to sharing models, which is doubly dumb because we want open source models, not closed source models. It’s a harmful status quo to try and enforce.