#AI generated sexual imagery may be labeled “deepfake,” but the violation can feel very real. A University of Washington #UWAllen UW CSE Security Lab Georgetown University study revealed varying attitudes to synthetic imagery—raising important questions around privacy + consent. #soups2024 washington.edu/news/2024/08/0…

Super great presentation from Natalie Grace Brigham at #soups2024, for her paper "'Violation of my body:' Perceptions of AI-generated non-consensual (intimate) imagery", also with Miranda Wei and Elissa M. Redmiles, Ph.D. . Paper at usenix.org/conference/sou…

And, super great talk from Natalie Grace Brigham on Sunday, too, on "Measuring and mitigating harms at the intersections of generative AI and image-based sexual abuse", at the #soups2024 Societal & User-Centered Privacy in AI (SUPA) workshop supa-workshop.github.io/supa.github.io…

At #SoLaR poster session (11 AM West Room 121,122) 1. Case Studies of Generative AI’s Use in Journalism arxiv.org/abs/2406.13706 2. Differentially Private Learning Needs Better Model Initialization and Self-Distillation arxiv.org/abs/2410.17566

Congrats to lab alumni Natalie *Grace* Brigham for presenting at the NeurIPS 2024 workshop on Socially Responsible Language Modelling Research (SoLaR)! See the poster: lnkd.in/g4KGbKum SoLar workshop: solar-neurips.github.io