AI-generated videos are about to improve dramatically with Google’s launch of the new Veo 3 video generation model and the AI filmmaking tool, Flow. Alongside these, Google is also rolling out several updates to the existing Veo 2 model.
At Google I/O 2025, amidst a wave of AI innovations, Veo 3 stood out to me. This cutting-edge video generation model incorporates real-world physics into your prompts and supports image inputs, promising smoother, more realistic videos without awkward movements or animations.
The best part is that Veo 3 can generate not just video but also accompanying audio. It can create sounds like “traffic noise on a city street, birds singing in a park, and even character dialogues,” all with nearly perfect lip-syncing, as demonstrated in this clip of the old sailor.
In addition, Google has enhanced Veo 3 with new features over its predecessor. Now, you can use images as prompts to include specific characters in your AI videos. You’ll also have control over the camera, allowing you to move it around while Veo 3’s AI fills in the gaps. Plus, it lets you effortlessly add or remove objects within the footage.
All these features are now part of Flow, Google’s new AI filmmaking tool. Developed in collaboration with artists, filmmakers, and content creators, Flow “lets you effortlessly create cinematic clips, scenes, and stories by combining Google DeepMind’s most advanced models: Veo, Imagen, and Gemini.“
Veo 3 is currently available to Google AI Ultra subscribers and Vertex AI enterprise users in the US, with plans to expand to more countries soon. Flow is now available for Google AI Pro and Ultra plan members and will be added to the Vertex AI API in the coming weeks.
