BYTEDANCE LAUNCHES OMNIHUMAN-1: AI THAT TRANSFORMS PHOTOS INTO LIFELIKE HUMAN VIDEOS

Summary from the TLDR Design newsletter

OmniHuman-1 by Bytedance is an AI model that transforms images and audio into lifelike human videos. It captures natural motion, expressions, and speech synchronization with impressive realism. Trained on 19,000 hours of video, OmniHuman-1 adapts to various formats. While not publicly available, its potential raises ethical and regulatory concerns.