At the Google I/O 2025 Conference, Google announced Veo 3, its newest step in AI video generation. Designed by DeepMind. Veo 3 allows users to generate realistic videos complete with synchronized audio, including dialogue, sound, and background noise, from just text or pictures.

What Is Veo 3?

• Integrated Audio Creation: Veo 3 can produce audio tracks that match the generated visuals resulting in a higher level of realism in AI-generated videos.
• Improved Realism: The model has produced improved motion dynamics, proper lip sync capabilities, and assembled narratives properly resulting in a more lifelike and cohesive videos overall.
• Higher Quality: Veo 3 is capable of supporting 4K video generation allowing creators to have the high-resolution content needed for professional content creation.

Veo 3 is available to consumers in the United States using Google’s AI Ultra subscription plan through the Gemini app (USD $249.99 a month). Enterprise users are also able to access Veo 3 through Google’s Vertex AI platform

Along with Veo 3, Google is also releasing Flow, an AI-powered filmmaking software to help and assist creative industry professionals in producing cinematic clips and scenes. Flow is able to use all the features of Veo 3 to provide advanced control to users in telling their stories.

To address concerns regarding the authenticity of AI-created content, Veo 3 will also be using SynthID watermarking, which adds invisible watermarks to video frames to help identify and authenticate AI-created media.

With these advancements, Google aims to empower creators and redefine the landscape of digital content production.