Glaze: The Digital Cloak Protecting Artists from AI Style Mimicry
How University of Chicago's Glaze tool uses adversarial perturbations to cloak artists' personal styles from AI image generators — a technical explainer.
Nightshade: How Data Poisoning Became a Weapon for Artists Against AI
A deep dive into Nightshade — the data poisoning tool from University of Chicago that lets artists fight back against unauthorised AI training by corrupting the models themselves.
Poisoning Datasets: A Theoretical Approach to Data Sovereignty
An analysis of techniques used to inject adversarial noise into public datasets to disrupt corporate scraping.