The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models. Is intended as a way to fight back against AI companies that use artists’ work to train their models without the creator’s permission.
ARTICLE - Technology Review
ARTICLE - Mashable
ARTICLE - Gizmodo
The researchers tested the attack on Stable Diffusion’s latest models and on an AI model they trained themselves from scratch. When they fed Stable Diffusion just 50 poisoned images of dogs and then prompted it to create images of dogs itself, the output started looking weird—creatures with too many limbs and cartoonish faces. With 300 poisoned samples, an attacker can manipulate Stable Diffusion to generate images of dogs to look like cats.
A large portion of AI art out there is made with Stable Diffusion, which can be run locally for free, and has a robust ecosystem of hobbyist trained models, LoRAs, etc. There are also somewhat competitive freely available LLM models.
Most attacks on AI that I see function as protectionism, where the biggest companies will end up being fine, but the people trying to do their own thing are the ones to be locked out.