Google introduces Flow, an AI video generator, which is tuned for filmmaking

At its ongoing I/O 2025 developer conference event, Google has introduced Flow, a new AI-powered tool built for filmmaking and creative video production. It brings together some of Google’s most advanced AI models: Veo 3, Imagen 4, and Gemini, to help creators produce high-quality cinematic scenes with minimal effort. Further, the company has added that Flow is built ‘by and for creatives’ and is currently being used by a small group of filmmakers to sketch out a lot of storyboard scenes by hand.

Google’s Flow AI video generation tool: Key details

Effectively an extension of the experimental Google Labs VideoFX tool, which was launched last year, Flow uses basic text-to-video prompts and a few images to build eight-second AI-generated clips. It then uses Flow’s Scenebuilder tools to stitch multiple clips together and bring cinematic visions to life using the power of artificial intelligence. It supports different ways of creating videos, including starting from text descriptions or adding shots around existing videos.

Flow also offers direct manipulation of camera angles, perspectives, and motion, with easy editing of scenes or widening up a shot to include more action. Additionally, it includes tools for organising assets and prompts. Google has also introduced something called Flow TV, which showcases multiple clips created with Flow, thereby allowing users to view shared projects, see the prompts used, and learn how different effects were achieved.