The startup behind Stable Diffusion has launched AI for video creation.
what happened: Generative AI startup Airport, which co-created last year’s text-to-image model Stable Diffusion, has released an AI model that transforms existing videos into new ones by applying a text query or reference image style.
What it does: In a demo reel posted on the site, the Gen-1 model shows how it can turn people on the street into clay dolls and books on a table into a cityscape at night. Other recent text-to-video models can generate very short video clips from scratch, but the Gen-1 adapts to existing footage and can produce very long videos.
Why is it important? Last year’s explosion in generative AI was fueled by millions of people getting their hands on powerful creative tools for the first time and sharing what they did, and Runway Gen-1 will have the same impact on generated videos. Read the full story.
– Douglas Garden
Why Finding AI-Generated Content Is So Hard (And What To Do About It)
Last week, OpenAI unveiled a text recognition tool powered by AI system ChatGPT. But don’t get too excited if you’re a teacher dreading the coming flood of chatgpty-generated essays.
