Today in Generative Media
GenAI for the least meaningful things; X begins AI ads; AI 'inspo' maddens hairstylists
News and Opinion
Generative AI is most useful for the things we care about the least (The Conversation)
X Rolls Out AI-Generated Ads in Push to Win Advertisers Back (AdWeek)
AI ‘inspo’ is everywhere. It’s driving your hairstylist crazy. (Washington Post)
Finally, something is puncturing conspiracy theories (Washington Post)
Drowning in "Untitled": Can Generative AI Save Us From Digital Repetition? (Jaclyn Konzelmann on Substack)
Hot take: GPT 4.5 is a nothing burger (Gary Marcus on Substack)
Software
ElevenLabs Launches Publishing Platform for AI-Generated Audiobooks (Publishers Weekly)
After rolling out Sora beyond the US, OpenAI plans to put the video AI tool inside ChatGPT (Yahoo! Tech)
AI SDK For Clipping On Mobile (Banuba) Let your users create video content easily. Boost session length and retention. No templates required.
Research
GHOST 2.0: Generative High-fidelity One Shot Transfer of Heads (project page)
GCC: Generative Color Constancy via Diffusing a Color Checker (project page)
Mobius Text to Seamless Looping Video Generation via Latent Shift (project page)
Distill Any Depth: Distillation Creates a Stronger Monocular Depth Estimator (project page)
NotaGen: Advancing Musicality in Symbolic Music Generation with Large Language Model Training Paradigms (project page)
Understanding Screenwriters' Practices, Attitudes, and Future Expectations in Human-AI Co-Creation (arXiv)
Dimitra: Audio-driven Diffusion model for Expressive Talking Head Generation (arXiv)
LPM: Efficient 3D Content Creation from Single Image by Large-Scale Partial 3D Modeling (IEEE Xplore)
Synthesizing Consistent Novel Views via 3D Epipolar Attention without Re-Training (arXiv)
The Impact of Generative AI on Work Productivity (Federal Reserve Bank of St. Louis)