Today in Generative Media

YouTube tests AI features; Adobe's genAI in the news; Studios align with Big Tech

• YouTube to test generative AI features, including a comments summarizer and conversational tool (TechCrunch)

• Adobe responds to controversy over AI-generated images of Gaza explosion (VentureBeat)

• As AI Battle Lines Are Drawn, Studios Align With Big Tech in a Risky Bet (Hollywood Reporter)

• PixArt-Alpha 1024px is a transformer-based text-to-image diffusion system trained on text embeddings from T5. This demo uses the PixArt-alpha/PixArt-XL-2-1024-MS checkpoint. (HuggingFace)

• Idempotent Generative Network (project page)

• Consistent4D: Consistent 360° Dynamic Object Generation from Monocular Video (project page)

• I2VGen-XL: High-Quality Image-to-Video Synthesis via Cascaded Diffusion Models (project page)

• Happy to release (the first?) AI Interactive Series on YouTube. (X)

• Using EverArt I can achieve amazing control of camera angles and while keeping character consistency. (X)

• "The constraints on storytelling are not going to be bound to budgets, who you know in the industry, how much knowledge you have about an obscure technique. It's about how good is your idea." (X)

• DALL E 3: Design 3D characters for your animated movies (X)