Today in Generative Media - 10/24/25
Wonder Studios gets $12M; Synthesia's realistic deepfakes; Curators, Sotheby's face AI art
News and Opinion
OpenAI-Linked Creative Studio Raises $12 Million as It Eyes AI Content Drive (Wall Street Journal)
An AI startup made a hyperrealistic deepfake of me that’s so good it’s scary (MIT Technology Review)
Amid the A.I. Deluge, What Counts as Art? Ask the Curators. (New York Times)
From slop to Sotheby’s? AI art enters a new phase (MIT Technology Review)
‘Legacies condensed to AI slop’: OpenAI Sora videos of the dead raise alarm with legal expert (The Guardian)
How platforms are handling the slop backlash (Platformer)
The platform exposing exactly how much copyrighted art is used by AI tools (The Guardian)
The Housing Market Is Already Terrible. A.I. Is Making It Even Worse. (Slate)
An Opinionated Guide to Using AI Right Now. (Ethan Mollick on Substack) What AI to use in late 2025.
Software and Data
VisoMaster is a powerful yet easy-to-use tool for face swapping and editing in images and videos (GitHub)
FlashWorld Spark Demo (HuggingFace)
LanPaint: Universal Inpainting Sampler with “Think Mode” (GitHub)
Turn your audio into sheet music and MIDI. (Songscription) Upload your audio file or YouTube link and let our advanced AI pick out the notes. Perfect for learners, composers, teachers, and producers.
Introducing ImagenWorld: A Real World Benchmark for Image Generation and Editing (ComfyUI blog)
Research
CompSlider: Compositional Slider for Disentangled Multiple-Attribute Image Generation (project page)
FLASHVSR Towards Real-Time Diffusion-Based Streaming Video Super-Resolution (project page) you can use it in comfyUI (GitHub)
3DPR: Single Image 3D Portrait Relight using Generative Priors (project page)
NANO3D: A Training-Free Approach for Efficient 3D Editing Without Masks (project page)