Cartwheel, a new 3D animation startup founded by former OpenAI scientist Andrew Carr and former Google creative director Jonathan Jarvis, seeks to make AI-generated animations more editable and controllable. The company focuses on converting simple 2D videos into precise 3D skeletons, allowing artists to refine outputs rather than relying on uneditable 'black box' prompts. Founders emphasize providing a 'control layer' to prevent AI sameness and enable open-ended storytelling.
Andrew Carr and Jonathan Jarvis launched Cartwheel to address limitations in current generative AI animation tools, where prompts often produce flawed results like distorted feet that are hard to fix without starting over. Their approach generates editable 3D data from 2D inputs, such as a backyard dance video turned into a realistic 3D skeleton. This lets creators adjust poses, lighting, and environments post-generation, treating AI as a tool rather than a final product replacement. “The output of our system is designed for people to edit. It's designed for people to touch and manipulate, and we don't want someone to type something in and then have it shuffle through to a finished animation. That's not the point of it. That's boring, who's going to watch that?” Carr said. The founders highlighted the scarcity of 3D motion data compared to abundant text, images, and videos used by big tech. “We knew it was going to be hard, but it turns out to be harder than we thought by probably a factor of 10 or 100 to get that data,” Jarvis noted. Cartwheel maps human movements to build nuanced performance models. Looking ahead, the company envisions AI powering real-time, reactive characters for gaming and social media, supporting open-ended world-building. Carr shared a key hypothesis: “everyone will work in 3D even if it's authored in 2D, even if the final output is just 2D video” within the next three years. By focusing below the pixel layer, Cartwheel aims to automate technical tasks while preserving human creativity in timing and taste.
