Conversation

Notices

1. Embed this notice
   Alex Stamos (alex@cybervillains.com)'s status on Thursday, 21-Dec-2023 11:32:04 JST Alex Stamos

   in reply to

   • David Thiel

   Lots of people have worried about CSAM in training sets, including the LAION team themselves, but David actually created a novel mechanism to detect it.
   Hopefully this will change the way these training sets are created in the future.

   @det

   • Embed this notice
     Alex Stamos (alex@cybervillains.com)'s status on Thursday, 21-Dec-2023 11:32:05 JST Alex Stamos

     • David Thiel

     How does Stable Diffusion 1.5 know how to create CSAM? It turns out it was trained on thousands of illegal images contained in the extremely popular LAION-5B image set.
     I’m so incredibly proud of my friend and colleague @det

     Story:
     https://www.404media.co/laion-datasets-removed-stanford-csam-child-abuse/

     Paper:
     https://stacks.stanford.edu/file/druid:kh752sm9123/ml_training_data_csam_report-2023-12-20.pdf

     Mike McCue repeated this.