Letter says many of works being sold by Christie’s are made by AI models trained on pieces by human artists, without a licence
Thousands of artists are urging the auction house Christie’s to cancel a sale of art created with artificial intelligence, claiming the technology behind the works is committing “mass theft”.
The Augmented Intelligence auction has been described by Christie’s as the first AI-dedicated sale by a major auctioneer and features 20 lots with prices ranging from $10,000 to $250,000 for works by artists including Refik Anadol and the late AI art pioneer Harold Cohen.
In principle all samplers are deterministic because they use PRNGs, any and all actual non-determinism you see is due to GPUs, underlying acceleration libraries playing fast+loose with numerical accuracy. Which they do because the models are generally robust against noise which is exactly what lacking numerical accuracy (or quantisation) is, but you can get into situations where, on direct comparison, it does make a difference.
You are not logged in. However you can subscribe from another Fediverse account, for example Lemmy or Mastodon. To do this, paste the following into the search field of your instance: !technology@beehaw.org
A nice place to discuss rumors, happenings, innovations, and challenges in the technology sphere. We also welcome discussions on the intersections of technology and society. If it’s technological news or discussion of technology, it probably belongs here.
Remember the overriding ethos on Beehaw: Be(e) Nice. Each user you encounter here is a person, and should be treated with kindness (even if they’re wrong, or use a Linux distro you don’t like). Personal attacks will not be tolerated.
Prompt, seed, model, LORAs, and better hope it’s a sampler that reliably produces the same results each time for the same input as not all of them do.
Nah, inability to produce the actual image is the point. All the “artist” did was type in a box, so that’s all the purchaser gets.
In principle all samplers are deterministic because they use PRNGs, any and all actual non-determinism you see is due to GPUs, underlying acceleration libraries playing fast+loose with numerical accuracy. Which they do because the models are generally robust against noise which is exactly what lacking numerical accuracy (or quantisation) is, but you can get into situations where, on direct comparison, it does make a difference.