GoogleThe company published a blog post yesterday (December 4) announcing a private beta preview on the Vertex AI platform thatOpen up the latest generative AI Video Model Veo, powering the enterprise content creation process.
Google initially publicly demonstrated the Veo model in May of this year, three months after OpenAI's Sora model, but Google's early release of Veo took advantage of the increasingly heated competition in the field of AI video generation.
1AI cites a blog post that describes Veo's ability to generate 1080p HD videos based on text or image cues, with a variety of visual and cinematic styles, although the initial release generated videos that were about a minute or more in length, although the blog post doesn't make clear the length limit of the videos that can be generated with the current preview.
From the example clips released by Google, Veo generates videos with such a high degree of realism - for example, the dog's fur pattern and collar in the video remain consistent during movement - that it's hard to tell they're being generated by AI if you don't look closely.