I thought I read some years ago about a supreme court ruling that "fictional" abuse material was legal because there is no real victim. Maybe that has been superseded by later rulings.
The ruling was that “fictional” abuse material was legal because it was an expression of free speech. Courts have thus far unanimously agreed, however, that generative AI doesn’t constitute authorship. Under the same principle, generative AI should not constitute protected speech under the 1st Amendment.
I don’t follow that logic. We would be in bad shape if we applied the standard of copyrightability to protected speech. By that logic, making a derivative Winnie The Pooh work would be unprotected speech on the grounds that it was public domain.
8x smaller! That’s pretty bonkers and awesome! I hope this will become the standard as it will allow the AI Horde workers to server 10x more models each and even faster (as it will cut down the loading times). I hope it doesn’t break Loras though.
Our commercial Stable Audio product produces high-quality, full tracks with coherent musical structure up to three minutes in length, as well as advanced capabilities like audio-to-audio generation and coherent multi-part musical compositions.
Stable Audio Open, on the other hand, specialises in audio samples, sound effects and production elements. While it can generate short musical clips, it is not optimised for full songs, melodies or vocals. This open model provides a glimpse into generative AI for sound design while prioritising responsible development alongside creative communities.
I wonder if after this release, people will train and release better models that aren’t just a gimped teaser of the pay product. That would be nice.
Whether or not you agree with AI image generation, the authors of this study have pulled off something impressive. This particular study isn’t going to be the single most important thing to humanity this year, sure, but they made a pretty clever stride in pushing a developing field forward and you don’t need to be excited about the field itself to appreciate that.
I’m assuming your dislike for AI image generation is based on the plagiarism issue, which is absolutely valid, but model architecture is separate from training data and the concepts here are perfectly usable with a more moral training set. The companies scraping all the data - OpenAI, google, and to a much lesser extent stability AI - are the ones to blame for that problem, not researchers working on model architecture.
I hadn’t, and it was definitely worth reading so thanks for the link. I’m still not sure exactly where I stand on the big AI companies relentlessly scraping everything for training data, but that was very convincing that copyright laws aren’t the solution (and I already believed better labour laws were needed for artists, though the details of exactly how music artists are getting shafted were new to me). Thanks for the link!
Big companies own large swathes of internet content already. The worst thing that could happen to us is new laws allowing them to band together into a data cartel that would keep us from benefitting from our shared world culture in the same way they exploit it for profit. We have everything to lose, because the second it becomes profitable, they’ll burn it all down, like they do with countless live service games, entire social networks, and movies.
Stable Diffusion is an open source image generating machine learning model (similar to Midjourney).
Stable Diffusion 3 is the next major version of the model and, in a lot of ways, it looks better to work with than what we currently have. However, up until recently we were wondering if we would even get the model since Stability AI ran out of funding and they’re in the midst of being sold off.
The “weights” refer to the values that make up the neural network. Basically by releasing the weights they are essentially saying that they are making the model open-source so that the community can retrain/fine-tune the model as much as we want.
They made a wait list for those who are interested in getting notified once the model is released, and they turned it into a pun by calling it a “weights list”.
I haven’t looked into the built in features in a while but I think comfy needs to have more basic nodes. I’ve picked up workflows with dead nodes and it’s not easy to see what used to be there and fill in the gap. Sometimes it’s something complex like an animation node but I’ve had basic text box nodes go dead.
stable_diffusion
Hot
This magazine is from a federated server and may be incomplete. Browse more on the original instance.