Google Veo 3 is the next generation artificial intelligence tool developed by Google DeepMind that allows users to create short, high-quality videos from simple text or image prompts. Full access to Google Veo 3 is since May 2025 available in the USA but it is expected to be available also in European countries soon.
What is new in Veo 3? Unlike its predecessors, Veo 3 can generate videos with synchronised audio, including dialogue, background music and sound effects, all integrated directly into the video. This means that users can input a descriptive prompt, and Veo 3 will produce a video clip that not only visualises the scene but also includes the appropriate sounds and voices. It creates realistic visuals, accurately simulating meaningful expressions, physical movements, lighting, and environmental details. It also maintains consistency in character appearance and behaviour throughout the video, which is something that was not great in previous models. These features make it particularly useful for creators, video makers, storytellers but also educators, who wish to produce content without extensive resources as scenery and effects can all be artificially created and added. Veo 3 will be accessible also in Europe through Google’s Gemini app and the Flow platform, with various levels of access depending on your subscription plan (AI Pro or Ultra). The new model shows great progress in creating realistic human videos, although it is not flawless and it can still misinterpret user instructions. Some technical limitations exist but Veo 3 and especially when paired with Flow, offers great new possibilities for storytelling. Watch this video on YouTube to get an impression of what Flow and Veo 3 can do. According to the creator, Hashem Al-Ghaili, this video was made in 8 hours’ time, making use of Veo 3, Eleven Labs, Suno AI and ChatGPT.
Image courtesy The Visitors (made with Veo 3) by Hashem Al-Ghaili
