Google has advanced its AI filmmaking tool, Flow, to produce videos that appear even more lifelike, making it challenging to immediately recognize them as AI-generated. On Wednesday, the tech giant revealed that users can now modify shadows and lighting in their AI-created videos. These enhanced editing capabilities are part of the Veo 3.1 update, also introduced on the same day, which Google claims improves the video creation process using images provided as prompts.

Flow’s latest features also enable users to incorporate audio into their videos. With the “Ingredients to Video” feature, users can produce videos with sound based on three reference images. Meanwhile, “Frames to Video” allows for the creation of a video that smoothly transitions from one image to another, complete with audio. Additionally, the “Scene Extension” feature offers the ability to extend the final second of a clip, generating up to an additional minute of video and audio.

The Veo 3.1 update retains the same pricing as its predecessor, Veo 3, and is accessible as part of a “paid preview” via the Gemini API for developers, as well as within the Gemini app.

Google has also announced an upcoming feature for Flow users, allowing them to remove any element from a video seamlessly. The tool will reconstruct the scene and background, ensuring it appears as if the removed object was never present.