This weekend in Generative Media

Vendors vs. AI nudes; Huang: Graphics needs AI; TuSimple's AI gaming pivot

News

• White House extracts voluntary commitments from AI vendors to combat deepfake nudes (TechCrunch)

• 'We can't do computer graphics anymore without artificial intelligence. We compute one pixel, we infer the other 32': Jensen thinks AI is integral to next-gen graphics tech (PC Gamer)

• Investors rebel as TuSimple pivots from self-driving trucks to AI gaming (TechCrunch)

• Hasbro CEO says all his mates are using AI for their D&D games, which is apparently 'a clear signal that we need to be embracing it' (PC Gamer)

• Blender artist has to prove their work isn't AI (Creative Bloq)

• Today we are announcing the formation of World Labs: a spatial intelligence AI company building Large World Models (LWMs) to perceive, generate, and interact with the 3D world. (worldlabs.ai)

• Here's our first look at the Google Photos extension for Gemini (APK Teardown) (Android Authority)

Software

• Gen-3 Alpha Video to Video is now available on web for all paid plans. (X)

• DepthCrafter: Generating Consistent Long Depth Sequences for Open-world Videos (GitHub)

Research

• AI can change belief in conspiracy theories, study finds (The Guardian)

• PoseTalk: Text-and-Audio-based Pose Control and Motion Refinement for One-Shot Talking Head Generation (project page)

• Gaussian Garments: Reconstructing Simulation-Ready Clothing with Photo-Realistic Appearance from Multi-View Video (project page)

• LT3SD: Latent Trees for 3D Scene Diffusion (project page)

• SwinGS: Sliding Window Gaussian Splatting for Volumetric Video Streaming with Arbitrary Length (arXiv) WebGL SwinGS Viewer

Misc

• It's like anyone can pick up a paintbrush, but that doesn’t mean they’ll paint a masterpiece. Tools don't replace taste, vision, or understanding how things fit together. (X)

• What do you do when video to video is really sticky to the structure of your input video? Distort and change the input video! (X)

• think really hard (X)