Autodesk Flow Studio’s new 3D Editor + Canvas solves one of the biggest limitations of AI video creation: precise control. Rather than relying solely on 2D video models that offer less control over the outcome, creators can build and direct scenes in a realistic 3D environment–giving them precise control over cameras, animation, blocking, and the environment–while still taking advantage of 2D video models for the final look.
As generative AI tools make it faster to create characters, environments, and other visual assets, the next challenge is turning those elements into cohesive shots that creators can direct, control, and refine. Creating assets is becoming easier but shaping them into intentional scenes still requires precise control over performance, camera movement, composition, and timing. Today, Autodesk announced 3D Editor + Canvas, a major expansion of Autodesk Flow Studio and the next step in its evolution into an AI-native creative platform. Rather than relying on prompts or reference images to direct characters, animation, camera movement, and composition through a 2D interface, creators can now work directly in 3D while AI handles the final rendering. In 3D Editor + Canvas, creators can then combine the characters, assets, and scenes created in Flow Studio with AI-generated worlds and environments from tools like World Labs Marble, making it possible to build fully realized cinematic scenes in minutes. Our focus on creative control isn’t new. For decades, Autodesk has powered the world’s most iconic films, television shows, and games with tools like Maya, 3ds Max, and Arnold, and we’re now extending that same production-proven approach into AI-powered storytelling with Flow Studio.
From generation to direction
We live in 3D, and we direct in 3D, yet most generative AI video tools still ask creators to play and edit in a 2D world. Trying to direct a performance or design a camera move by prompting or annotating a flat image is inherently limiting. With 3D Editor, creators can block scenes, direct performances, and control timing, pacing, and camera movement in 3D. Then, in Canvas, they can generate, refine, and shape the final look in 2D. Together, 3D Editor + Canvas gives users the flexibility to stay in 2D when they want speed and simplicity or move into 3D when they need more precise control over composition, camera motion, and scene construction. It brings the strengths of both spaces together while letting creators choose how much control they need for each shot.
3D Editor gives creators precise control to stage assets, design scenes, and direct cinematic shots in Autodesk Flow Studio.
Why this matters now
At this year’s Cannes Film Festival, the conversation around AI filmmaking focused on authorship, creative intent, and control. As AI becomes more capable of generating characters, environments, and visual assets, the challenge is no longer about simply creating content–it’s about how to direct it. That’s exactly the problem we’re focused on solving with 3D Editor + Canvas–we want to give filmmakers, artists, marketers, studios, and storytellers greater creative control while removing technical and financial barriers to creating their work.
The evolution of Flow Studio
By combining generation, direction, scene assembly, rendering, and refinement within a single platform, Autodesk is advancing its vision of Flow Studio as an AI-native creative platform built for the next generation of filmmakers, studios, artists, marketers, and storytellers. 3D Editor + Canvas also establishes the foundation for potential future capabilities, including expanded Autodesk Maya and other DCC integrations, additional asset import options, AI-assisted scene construction, and agent-driven directing workflows. The future of AI filmmaking won’t be defined by the tools that generate the most content. It will be shaped by the tools that give creators the most control, preserving creative intent from idea to final frame.
