⚡ NEW: En/tag/omni - Collection

Published on 2026/06/01
GEMINI OMNI AND THE FUTURE OF VIDEO

Over the last few years, we've seen artificial intelligence systems capable of generating images from scratch. Then came the ones that could create entire videos from a simple text prompt. And now it looks like we're entering a new phase.

Google has unveiled Gemini Omni, a tool capable of modifying already recorded videos simply by talking to it.
And when I say talking, I literally mean writing instructions as if you were speaking to another person. "Open the blinds." "Turn on the lights." "Replace that object with another one." "Add an animal next to the bed." "Change the camera angle."

The AI analyzes the original video and makes the requested changes while maintaining visual consistency, respecting the lighting, perspective, movement, and much of the scene's physics.
What makes it truly remarkable is that it isn't creating a completely new video from scratch. It's working on an existing recording and transforming it on the fly according to the user's instructions.
That means many tasks that used to require hours of editing, retouching, visual effects work, or even reshooting scenes could now be solved simply by describing what you want to happen.

And this is only the beginning. According to Google, the version shown uses the Flash model, a lighter and faster version of the technology. In other words, what we're seeing today will probably look far less impressive than what we'll be seeing in a year or two.
The implication is obvious. In the near future, cameras may stop capturing the final version of a scene. They may only capture the first draft. And then artificial intelligence will finish the job.




# Watch videos




ExtraBall
They're completely out of control.

instagram.com/amahia_franco




ExtraBall2