Google wants to make it easier to create AI-generated videos, and it has a new tool to do it. It’s called Flow, and Google is announcing it alongside its new Veo 3 video generation model, more controls for its Veo 2 model, and a new image generation model, Imagen 4.
With Flow, you can use things like text-to-video prompts and ingredients-to-video prompts (basically, sharing a few images that Flow can use alongside a prompt to help inform the model what you’re looking for) to build eight-second AI-generated clips. Then, you can use Flow’s scenebuilder tools to stitch multiple clips together.
Flow seems kind of like a film editing app, but for building AI-generated videos, and while I’m not a filmmaker, I can see how it might be a useful tool based on a demo I saw. In a briefing, Thomas Iljic, a product manager at Google Labs, showed me a few examples of Flow in action.
In one demo, we watched an animated-style video; the “camera” zoomed out to reveal that the video was playing on a TV; the video zoomed out again to show the room the TV was in. Then, the “camera” slowly flew through a window and watched a truck pass by.
It all looked pretty seamless, though I was only briefly seeing the video in a tiny Google Meet window, so I can’t speak to any AI strangeness that might be visible if you look closely. But the idea for Flow isn’t so much about creating long videos. Instead, it’s more about helping filmmakers quickly get their ideas “on paper,” says Iljic.
As for Google’s new models announced at I/O, Veo 3 will have better quality, is easier to prompt, and can generate video and sound together (including dialogue), Matthieu Lorrain, creative lead at Google DeepMind, tells The Verge. It’s also better at understanding longer prompts and correctly handling a succession of events in your prompt.
Veo 2 will offer tools like camera controls and object removal. And Google’s new image generation model, Imagen 4, has improved quality, can export in more formats, and is apparently better at writing real text instead of the AI garble that often appears in these images.
Flow is launching today in the US for people who subscribe to Google’s new Google AI Pro and Google AI Ultra plans. “Google AI Pro gives you the key Flow features and 100 generations per month, and Google AI Ultra gives you the highest usage limits and early access to Veo 3 with native audio generation,” according to a blog post.
Read the full article here
