Summary
AI video platform Morphic has launched 3D Scene, an editor that lets creators block characters, movement, cameras and lighting in a virtual set before generating final footage. A conversational copilot can assemble an initial scene, after which users directly adjust the 3D layout. Morphic says the tool includes about 2,500 assets, more than 200 adjustable actions and support for crowds of up to 60 characters. Its Cast by Color system assigns each mannequin a color and name linked to the generation prompt, aiming to preserve positions, direction and eyelines in multi-character shots. Creators can choose twelve camera-movement types, including drone and crane moves. Automated checks flag violations of the 180-degree rule and implausible movement speeds. Morphic says previsualisation does not consume paid generation credits, potentially reducing rerolls, but the saved report provides no independent cost or consistency tests.
Hana’s perspective
I see this as the right direction for AI video: move control upstream instead of asking creators to reroll expensive outputs. Blocking a shot in 3D gives directors a shared visual plan and makes camera, staging and continuity choices explicit. The color-cast system is a pragmatic bridge between spatial intent and models that still struggle with multiple characters. The unanswered question is fidelity—whether the generated footage consistently preserves the approved block, performances and style. Independent production tests should measure that before studios count on the promised savings.
Are you a writer with a great story? We'd love to hear from you: APPLY HERE