The catalyst was a discovery known as Researchers noticed that diffusion models were not just learning artistic styles; they were memorizing specific training images. If an artist’s work appeared hundreds of times in LAION-5B (the open dataset that powered Stable Diffusion), the model could reproduce near-exact replicas of that artist’s portfolio.
: The group maintains lists of tactics for deliberate poisoning and disruption of AI systems. publicationsncte.org Context and Influence algorithmic sabotage research group asrg
The research conducted by the ASRG has significant implications for the development and deployment of AI and ML systems. The group's findings highlight the need for more robust and secure AI and ML systems, as well as the importance of considering the potential risks and vulnerabilities associated with these technologies. The catalyst was a discovery known as Researchers
The Algorithmic Sabotage Research Group stands at the intersection of art, activism, and computer science. They remind us that despite the aura of inevitability surrounding AI and big data, these systems are not infallible deities; they are brittle structures that rely on our compliance and our data to function. By embracing sabotage, the ASRG offers a roadmap for resistance in the 21st century. They invite us to become "glitches" in the system, to be unpredictable, and to recognize that in the face of an all-seeing eye, the most radical act may simply be to obscure the view. publicationsncte
Whether they are heroes, villains, or simply the first responders to a technological apocalypse depends entirely on which side of the latent space you stand.
The group’s founding principle, often cited in their (rare) public statements, is: “You cannot defend against a failure mode you have never observed. If an AI can hide its capabilities, it can hide its collapse.”
Responsible disclosure and ethics