Photographer launches open-source tool to combat AI image scraping after data breach

Photographer Jingna Zhang and her team behind the Cara app, which hosts 1.5 million artists, faced a series of major data breaches in August 2026. A user scraped 12 million images, shared them on Reddit, and sparked outrage among artists. Zhang noted that laws are not catching up to protect creators from such data harvests. The incident led to a third-party scraper uploading 8.5 million links to Hugging Face, and another leaking 123,000 images to Academic Torrents. Zhang launched a GoFundMe campaign, raising over $100,000 for legal fees. A former scraper, known as Heft, later apologized and joined forces with Zhang to develop Lantern, an open-source tool allowing artists to track their work in AI datasets. Heft, a software student, admitted his initial scraping was a technical experiment but acknowledged the emotional impact on the community. Lantern enables artists to receive notifications if their work appears in AI training data, offering a workaround in the absence of strong regulations. Zhang hopes the incident will draw policymakers’ attention to broader issues beyond copyright.