Today in Generative Media

Newspapers want $$ for AI; AI for PowerPoint, YouTube Music genAI makes playlist art

• Newspapers want payment for articles used to power ChatGPT (Washington Post)

• Of course AI is coming for PowerPoint now (FastCompany)

• YouTube Music will let you make custom playlist art with gen AI (9to5Google)

• Zero123++: a Single Image to Consistent Multi-view Diffusion Base Model (GitHub). Demo (HuggingFace). Paper (arXiv).

• TexFusion: Synthesizing 3D Textures with Text-Guided Image Diffusion Models (project page)

• FreeNoise: Tuning-Free Longer Video Diffusion via Noise Rescheduling (project page)

• Localizing and Editing Knowledge in Text-to-Image Generative Models (arXiv)

• Matryoshka Diffusion Models (arXiv)

• My favorite AI moments are when I get to appreciate how natural & intuitive my exploration of an idea is. (X)

• RunwayML just hosted the GEN48 competition and the entries are phenomenal. (X)

• In-browser TTS in 3 lines of code! 🤯 (X)

• There are a large number of papers coming out which show that you can use LLMs to create realistic or interesting motions (X)

• You can make ChatGPT write like you by using the Code Interpreter (X)