Google dropped Gemini Omni Flash at I/O this week and it does something genuinely new: you describe a video, it generates it, you keep talking, it keeps updating. The whole conversation is the edit suite.
For brands that have been sitting on shaky phone footage because they can't afford a shoot, this is worth paying attention to.
What it is (without the press release language)
It's a video generation model that works like a conversation. You describe what you want. It renders it. You say "make the lighting warmer" or "add rain" or "now put that same scene inside a glass sphere" and it builds on the last output rather than starting from scratch.
That last part is the bit that matters. Most AI video tools treat every prompt as prompt one. This one has context. You're not babysitting a generator, you're having a back-and-forth with it.
What the prompts actually look like
We ran a few from the Google blog to put in the carousel on our Instagram. Real prompts, unedited:
"Claymation explainer of protein folding, everything is made out of clay, no hands, stop motion, accurate."
"When the person touches the mirror, make the mirror ripple beautifully like liquid, and the person's arm turns into reflective mirror material."
These are 10-second clips. Both rendered without a single asset, a single piece of footage, or a brief sent to a motion designer. See the full carousel on our Instagram.
What this means if you're running an ecommerce brand
Three things, in order of how quickly you can act on them:
- Product demos without a shoot. Describe your product in a scene. Adjust the lighting, the setting, the energy. No props, no photographer, no three-week turnaround on edits.
- Seasonal content at volume. Valentine's version, Mother's Day version, winter version. Describe the same clip with different context and render the variants. What used to be a production decision is now a prompt.
- Social content that doesn't look like stock. The reason AI-generated content has a tell is that most people use the same tools the same way. Gemini Omni's conversational editing means you can push scenes further than a single prompt would take them. The results start looking specific instead of generic.
Where to find it
It's rolling out now. YouTube Shorts and YouTube Create are free. The Gemini app has it on Plus, Pro, and Ultra plans. API access is coming for developers.
Go try it before your competitors do.
Sources