Google launches Veo to compete with OpenAI’s Sora

Google has unveiled Veo, which it believes is the most “capable video generation model” and main rival to OpenAI’s Sora.

Google announced several AI-powered novelties during its I/O developer event, including Veo, Google’s “most capable video generation model.”

The tool generates 1080p resolution videos that go “beyond a minute” in various cinematic styles and effects, including aerial shots and time lapses.

However, Google said that new features will be available for “select creators” in upcoming weeks.

The company also revealed that some of Veo’s features would be accessible in YouTube Shorts and other undisclosed products.

The US tech behemoth claims the video generation model was created to “make video production accessible to everyone.” Actor and filmmaker Donald Glover and his creative team, Gilga, revealed insights into how Veo works in reality by implementing it to create different scenes.

Veo’s generative video model can interpret any given text and connect this information with “relevant visual references,” the company said. Since the tool can understand natural language and visual information, it supposedly provides more detailed scenes.

In addition, Veo comes with an editing feature that allows users to add details found in another image. With the new tool, creators can also get consistent videos that are not disrupted by flickering objects between frames.

Google’s Veo has already been recognised as a competitor to OpenAI’s Sora – an AI tool that turns text descriptions into realistic and detailed video clips.