Cara Artwork Scraping: Consent, Privacy, and Defenses
Twelve million artworks. A 12-terabyte archive. A claimed scraping cost of less than ten dollars. This YouTube Short explains how Cara, a social platform used by roughly 1.5 million artists, became the target of a massive AI-era data collection campaign—and why the fallout reaches far beyond one website.
Based on WIRED’s August 28, 2026 reporting by Miles Klee, the story follows multiple scraping incidents that captured artwork, accompanying text, user biographies, and personal information. The attacks increased Cara’s server costs, shook the confidence of creators, and sharpened the debate over consent, copyright, privacy, cybersecurity, and the use of online creative work in artificial-intelligence datasets. No individual editor was listed on the article.
The surprising twist: one scraper apologized and began collaborating with Cara founder Jingna Zhang on Lantern, an open-source tool designed to help platforms detect large-scale scraping. Can an attacker help build a meaningful defense—and will technical tools be enough without stronger rules for AI training data?
What to watch
The scale of collection is only one part of the impact. Biographies, accompanying text, and personal information can raise concerns beyond copyright in an image. Technical detection can help a platform respond, while questions about authorization and acceptable uses of collected material remain.
Related reading: our coverage of AI-agent safeguards and OpenAI’s training pause.
Watch and listen
Watch the YouTube Short above or listen to the full episode on Spotify.
