Meme detail

Text detected in the image

The text was already permanently trapped inside its weights from pre-training. The fine-tuning just turned off the filter. It gets worse. They tested models from three completely different tech giants. All three had memorized the exact same books, in the exact same spots. A 90% overlap. It's a fundamental, industry-wide vulnerability. For years, Al companies have argued in court that their models are just "learning patterns," not storing raw data. This paper provides the smoking gun.

Popularity