At the end of 2024 Google presented see2, a model of Creation of videos with AI Very powerful that even activated Gemini users in Spain.
This new model was equivalent to what Openai had shown with Sora weeks before, but that was not yet in the hands of the end users.
Google knew that he had a great asset there and has played it. And now the bet on the Google I/or 2025 has uploaded with the announcement of its successor, I see3, capable of making better videos and putting sounds.
This new service is integrated into Google One’s most powerful ratesthose that give access to Google’s best tools. Some users, even some filmmakers, are already creating proposals.
In fact, the Being able to include not only sounds, but dialogues Perfectly usable, leaves us pieces like the next. In it the “actors” laugh at the idea of being part of an AI.
Prompt Theory (Made with I see 3)
What if ai-generated Characters refused to Believe they were ai-generated? pic.twitter.com/gjsrcxp3JP
-Hashem al-Ghaili (@hashemghaili) May 22, 2025
In addition to controlling what the characters say, and the situation, it is also possible to control the aesthetics of the videos.
In this creation, it can be seen how the intention was create something more similar to what we would see in a science fiction movie or in a Black Mirror chapter.
I Did More Tests With Google’s #I see3. Imagine If ai Characters Became Aware they were living in a simulation! pic.twitter.com/nhbrnqmtqv
-Hashem al-Ghaili (@hashemghaili) May 21, 2025
But the possibilities go much further. If stock videos already had a fearsome rival, now that also reaches other types of industries.
A user has spent $ 500 on making an advertisement on a drug convincing enough to consider whether it is viable to be used.
I USED TO SHOOT $ 500K Pharmaceutical Commercials.
I Made This for $ 500 in I see 3 credits in less than a day.
What’s The Argument For Spending $ 500k Now?
(Steal My Prompt Below 👇🏼) pic.twitter.com/4uh43exdux
– PJ Ace (@pjaccetturo) May 22, 2025
All these creations have come out in the early hours of availability of I see 3. In addition, Google presented Flow, a video editing system that promises a lot.
They can Make sequences of all kinds expanding and cutting clips of a few seconds, choosing the characters, and writing what we want What happens.
CREATED WITH GOOGLE FLOW.
Visuals, Sound Design, and Voice Were Prompted Using I see 3 text-to-video.
Welcome to a new was of filmmaking. pic.twitter.com/e3nsa1wsxe
– Dave Clark (@diesol) May 21, 2025
The upper video is a good example of what allows the new creation of the Mountain View giant. It is not only quality, but consistency.
In addition, it is not only possible to make photorealist videos, but Animation scenes can be created that seem taken from a professional film stid.
My first see 3 gen
> A video with dialogue of two muffins while baking in an over, The First Muffin Says “I Can’t Believe This I see 3 Thing Can Do Dialogue Now!”, The Second Muffin Says “AAAAH, to Talking Muffin!” pic.twitter.com/va2vuzf8ss
– Fof (@fofrai) May 20, 2025
But The most striking is the speed at which Google is improving its ia creation systems and of another kind. In a few years this will seem outdated at unthinkable levels.