AI Art Scraper Turns Collaborator on New Protection Tool
Cara, an art portfolio platform designed to shield creators from AI training scrapes, faced three major data grabs in August 2026, including a 12-terabyte archive of 12 million images posted on Reddit. The individual responsible later began collaborating with Cara's founder on a tool to help artists. The platform's founder, Jingna Zhang, is also involved in class-action lawsuits against AI companies over copyright infringement.
Cara’s founder, Jingna Zhang, has been fighting AI copyright battles on multiple fronts, including two class-action lawsuits against major AI companies. The platform’s protective tools, like Glaze, aim to disrupt style mimicry, but scrapes remain technically difficult to block. The first scraper’s dataset, posted on Reddit, contained nearly all of Cara’s public images, yet cost under $10 to compile. Subsequent attacks pulled metadata and personal bios, raising privacy concerns beyond just artwork. Zhang’s GoFundMe, nearing its $120,000 goal, will fund legal strategies under cyber and copyright law.
This incident could reshape how artists view platform security and legal recourse against data harvesting. If scrapes continue despite protective measures, creators may lose trust in even “safe” havens, pushing them toward more drastic technical or legislative solutions. The collaboration between a former scraper and an artist suggests potential for redemption and cooperative fixes, but the broader impact hinges on whether courts clarify scraping legality—affecting millions of artists and the AI industry’s data practices.