Google knows that every moment once carried its own sound, and the world was full of them.
But it's that natural harmony between sight and sound that made stories feel alive. For people, blending these senses came effortlessly. But for machines, it had long been a complex challenge. But when Google introduced Veo 3, it changed everything.
And later, when Veo 3.1 came, Google further refined this harmony.
With Veo that gets better, Google is stepping close to giving machines not just eyes, but ears that understood the world’s quiet music.
And this time, Google is updating Veo 3.1, to make it even better.
According to Google in an updated Veo 3.1 webpage:
Veo is getting new precision editing capabilities that let you easily add or remove elements from a scene - all while preserving the integrity of your original video. pic.twitter.com/VWvYDiqjMb
— Google DeepMind (@GoogleDeepMind) October 20, 2025
The first feature is allowing users to add new elements to any scene.
With 'insert,' users can introduce pretty much anything they have in mind, and have Veo 3.1 generate that imagination.
From realistic details to some fantastical creatures, according to Google, "Flow now handles complex details like shadows and scene lighting, making the addition look natural."
Add last-minute props or characters
Veo lets you place new elements into scenes and automatically handles all the complex details - from realistic shadows to environmental interactions - making the additions look completely natural. pic.twitter.com/h52Axci1O4— Google DeepMind (@GoogleDeepMind) October 20, 2025
The second addition, is the ability to remove unwanted objects.
Since AI-generated content may introduce unwanted objects or undesired elements, the update to Veo 3.1 allows users to remove these things. Or, if users wish to remove something for their own video, Veo 3.1 can also do that.
According to Google, users should soon be able to take anything out of a scene, and that "Flow will reconstruct the background and surroundings, making it look as though the object was never there."
Want to remove something from your scene?
Quickly erase unwanted objects while Veo reconstructs the background to keep the rest of your work untouched. pic.twitter.com/TKkviBk32W— Google DeepMind (@GoogleDeepMind) October 20, 2025
Veo 3 was introduced as Google’s bold answer to OpenAI’s Sora, a direct rival in the race to master AI-generated video.
But when Sora 2 arrived, it didn’t just compete. It dominated the conversation. Despite Veo 3 outperforming Sora 2 in several technical benchmarks, OpenAI’s model captured far more public imagination. Its distinctive, almost cinematic approach to video generation made it go viral in ways Veo 3 never quite achieved.
Sora 2 may not share the sharp wit or irreverent personality of xAI’s Grok, yet as a dedicated app, it pushed deepfake to a whole new level.
However, Sora 2 doesn't really blur the line between synthetic creation and lived reality the way Veo 3 can, at least in terms of realism.