Finding AI-Generated Faces in the Wild
A top-performing approach for detecting both GAN- and diffusion-generated faces in the wild.
Last year, the LinkedIn Anti-Abuse team developed a simple, novel approach that can detect 99.6% of StyleGAN-generated profile photos, and we published our first paper on AI image detection (more detail).
In the year since we released our first paper, this field of study and our own research have remained dynamic and fast-paced. We are excited to share our latest research paper, Finding AI-Generated Faces in the Wild, where we introduce a new concept for a model that can detect AI-generated images produced by a variety of generative algorithms. This new concept can tell the difference between real profile photos and those generated by different types of AI models, such as GAN-based StyleGANs, generated.photos, and EG3d, as well as diffusion-based Stable Diffusion, DALL-E 2, and Midjourney.
Our approach is detailed in the full paper that was released at the Workshop on Media Forensics at the 2024 Conference on Computer Vision and Pattern Recognition (CVPR) on June 18th 2024.