Today in Generative Media
Artists try Sora; NVIDIA CEO on AI for games; SAG defines performers as humans
Sora: First Impressions. We have gained valuable feedback from the creative community, helping us to improve our model. (OpenAI blog) OpenAI just gave artists access to Sora and proved the AI video tool is weirder and more powerful than we thought (TechRadar)
NVIDIA CEO: Soon there’ll be games where AI generates “every pixel” (Futurism)
Actors' Union Contracts Define Voice Performers Exclusively As Humans (PCMag)
Why watermarking won’t work (VentureBeat)
Large Language Models’ Emergent Abilities Are a Mirage (Wired)
The Online Degradation of Women and Girls That We Meet With a Shrug (New York Times)
Generative AI could leave users holding the bag for copyright violations (Yahoo News)
BBC Will Stop Using AI For ‘Doctor Who’ Promotion After Receiving Complaints (Deadline)
"AI native" Gen Zers are comfortable on the cutting edge (Axios)
Machine 'unlearning' helps generative AI forget copyright-protected and violent content (TechXplore)
Can you hear me now? AI-coustics to fight noisy audio with generative AI (TechCrunch)
ReNoise: Real Image Inversion Through Iterative Noising (HuggingFace)
Be Yourself: Bounded Attention for Multi-Subject Text-to-Image Generation (project page)
GeoWizard: Unleashing the Diffusion Priors for 3D Geometry Estimation from a Single Image (project page) Demo on HuggingFace
Champ: Controllable and Consistent Human Image Animation with 3D Parametric Guidance (project page)
We sadly found out our CTM paper (ICLR24) was plagiarized by TCD! (X)
the outcomes that occur when you give everyone access to something are nowhere near equitable. (X)
Open AI not open, stability ai not stable. Only MidJourney remains mid. (X)