This weekend in Generative Media

Russian intel and political deepfakes; Sci-fi got AI wrong; Open source for enterprise

News and Opinion

• American creating deepfakes targeting Harris works with Russian intel, documents show (Washington Post)

• Our Robot Stories Haven’t Prepared Us for A.I. (New York Times)

• The enterprise verdict on AI models: Why open source will win (VentureBeat)

• Architects are proving that AI is best used for the boring stuff (Fast Company)

• Adobe execs say artists need to embrace AI or get left behind (The Verge)

• After selling Anchor to Spotify, co-founders reunite to build AI educational startup Oboe (TechCrunch)

Software

• More transparency for AI edits in Google Photos (Google Keyword blog)

• Napkin turns your text into visuals so sharing your ideas is quick and effective

• Mochi 1 Preview now on RunPod (X)

• Act-One has been deployed to everyone. Refresh and enjoy (X)

Research

• LargeSpatialModel: End-to-end Unposed Images to Semantic 3D (project page)

• SMITE: Segment Me In TimE (project page)

• 3D-Adapter Geometry-Consistent Multi-View Diffusion for High-Quality 3D Generation (project page)

• Framer: Interactive Frame Interpolation (project page)

• Unbounded: A Generative Infinite Game of Character Life Simulation (project page)

• CtrLoRA: An Extensible and Efficient Framework for Controllable Image Generation (arXiv) Code on GitHub

• nvTorchCam: An Open-source Library for Camera-Agnostic Differentiable Geometric Vision (arXiv) Code on GitHub

Misc

• The "Winds of Yawanawá: Fabulous Triptychs," by Refik Anadol, beautifully merges AI and generative art with Yawanawá heritage and symbolism. (X)