You should read every single story about adverserial filters that break picture-bots with extreme scepticism. Remember dazzle facepaint that totes broke facial recognition? You know, that story that's still occasionally recycled, despite the fact it only ever worked against specific versions of specific facial recognition software, and was inevitably obsoleted by the fact of its own publicity?
Magic DRM is not going to save the world from shite generators.
(Also: whatever you do to artwork you're uploading now can obviously only affect affect some hypothetical future bot trained on a hypothetical future data set that includes your poisoned work. Y'know, assuming the adverserial filter is any more than a placebo against whatever specific version of whatever specific software that is.)
The only effective way to stop the deliberate proliferation of truth-eroding shitmachines is to address the incentives to their proliferation
string up a venture capitalist today
While investigating the StableDiffusion technology for an unrelated scholarship venture, I learned that it can't read the image data from PNG and JPG files exported by Krita. It was pretty cool to see it in action, but I do wonder about the potential work-arounds that may arise. Machine generated imagery systems are really quite sophisticated and only seem to be growing in complexity and capacity, and despite the core of the problem being venture capitalism, there is a larger picture that a game of Suits and Nooses simply will not be capable of resolving:
Internationally, we're approaching a Scifi Vortex of Machine Algorithm piloted software designed to attack infrastructure. Troll farms and various attack vectors employed by the Chinese, Russian, and U.S. intelligence services as part of global competition are increasingly investigating the capacities of automated self-learning machine algorithm systems as a way to attack each other. We might be approaching the part of a net-runner / cyberpunk style story where the "AI Wars" begin happening in the information infrastructure that connects the world. The petty small-scale profit motives of discount mercantilism be damned, so called Great Power competition is fundamentally pointed at much more than just wealth extraction, though said hoard of treasure is certainly a side effect.
I think what maybe deserves to be said with the above in mind is that there are countless actors who are involved in the many stakeholders who are investing in or innovating within the so-called AI / Neural Network / Machine Learning spaces. These agendas range from the mundane profit seeking to ambitious political goals, some are heady and are straight up motivated by a religious rejection of the null hypothesis in regards to the 'true' intelligence of these systems, thinking that they've invented god (actual shit I have read). Because of these dynamic motivations, it is hypercritical that people unlearn the lesson of "beating AI Art" and other such petty victories, because these technologies are going to become fundamental in the upcoming decades in regards to how virtually all systems of information infrastructure will be designed.
I think a lot about the meme of Billy May's being pulled from the grave and narrating the perils of voice deepfakes. I think this line (and the larger synthetic voice clip) illustrates my final thoughts, and I know people have fallen for this already since as far back as 2022 when Russia began phase 2 of invading Ukraine.
"Never trust your eyes or ears again in this modern digital hellscape"