Today in Generative Media

AI fears in Hollywood; Sora predictions; 60% of GPT-3.5 outputs are plagiarized

• AI’s New Job? All-Purpose Hollywood Crewmember (Hollywood Reporter)

• Tyler Perry Puts $800M Studio Expansion on Hold After Seeing OpenAI’s Sora: “Jobs Are Going to Be Lost” (Hollywood Reporter)

• What two years of AI development can tell us about Sora (Vox)

• OpenAI’s Sora text-to-video tool's impact will be ‘profound’ (ComputerWorld)

• Copyleaks Research Finds Nearly 60% of GPT-3.5 Outputs Contained Some Form of Plagiarized Content (Copyleaks blog)

• Run your ComfyUI workflow on Replicate (replicate.com) I'm enjoying that I can now quickly turn any ComfyUI workflow into a Replicate model. (X)

• Snap Video: Scaled Spatiotemporal Transformers for Text-to-Video Synthesis (project page)

• Consolidating Attention Features for Multi-view Image Editing (project page)

• T-Stitch: Accelerating Sampling in Pre-trained Diffusion Models with Trajectory Stitching (project page) Both Sora and Stable Diffusion 3 adopt diffusion transformers, but do we really need a super large DiT for all sampling steps for generation?🧐 No🙅‍♂️. We found ~40% early timesteps of DiT-XL can be replaced with a 10x faster DiT-S without image quality drop! (X)

• China's first AI anime will soon be aired on China's CCTV station. (X)