Making a video has always involved a little bit of theater. There is the writing, of course, but also the awkward setup: finding a quiet room, turning toward the light, checking whether the camera is level, recording the same sentence three times because the first two sounded strange. Then comes the editing — the part that can turn a 30-second update into an hour-long chore.

Google’s latest updates to Vids are aimed squarely at that friction. The company is adding Gemini Omni, a tool for generating and editing video through ordinary written prompts, alongside personal avatars that let users create a digital version of themselves to deliver a message without stepping in front of a camera.

Together, the features point to a familiar but increasingly consequential idea in workplace software: video should be as easy to produce as a document or slide deck. The question is whether making it easier also changes what, exactly, people are willing to put on camera.

Google Vids, the company’s AI-assisted video creation tool, already sits inside the broader Workspace world of Docs, Slides and Drive. Its new Gemini Omni integration pushes the product further into generative video territory. A user can start with a text description, add image references such as a product photo or rough sketch, and ask the system to create a clip around that combination.

The ambition is less “make a blockbuster” and more “help me get this idea out of my head.” A manager preparing a quick internal announcement, for instance, could describe a scene, supply an image of a new product and generate a concise explainer without opening a conventional editing timeline. That is a useful shift because much of business video does not fail for lack of creative vision. It fails because people run out of time before they ever press record.

What may matter even more is what happens after the first draft. Generative tools are often impressive at the moment of creation but frustrating when users want to make precise changes. A clip can look almost right, which is another way of saying it is not right at all. Google says Gemini Omni will allow people to refine generated clips — and even videos shot on a phone — with conversational requests. Swap the background. Brighten the scene. Add an effect. Keep working step by step, rather than beginning again from a blank prompt.

That sounds simple, but it addresses one of the least glamorous parts of video work: revision. Editing software has traditionally rewarded people who understand tracks, layers, keyframes and color controls. Those tools remain powerful, and they are not going away for professionals. But for the much larger group of people who simply need a clear, usable video by the end of the afternoon, describing an edit in plain English could be a far more natural interface.

The other update, personal avatars, takes the idea in a more personal direction. Users can upload a selfie and a short voice recording, then type a script for an avatar designed to look and sound like them. In theory, it means a person can send a video update, greeting or presentation without arranging a recording session at all.

There is a clear appeal here. Anyone who has tried to record a polished 60-second message knows how disproportionately difficult it can be. The camera can be inconvenient; the speaker may be traveling, tired, pressed for time or simply unwilling to appear on screen. An avatar offers a shortcut through all of that. It could be particularly handy for routine communications: training reminders, project updates, onboarding messages, localized announcements or a quick note to a distributed team.

But it also introduces a new kind of social calculation. A video feels more direct than an email in part because viewers assume they are seeing a person speak in that moment. An avatar blurs that expectation. The message may still be genuinely authored by the person whose face and voice it uses, but the performance is synthetic. For some audiences, that may be entirely fine. For others, especially in sensitive or high-stakes settings, it could feel less like a thoughtful video message and more like a polished substitute for one.

Google appears aware of that tension. The company says its personal avatars are tied to the account holder’s likeness and restricted to use by that account holder, a safeguard intended to limit impersonation. It is also attaching an invisible SynthID digital watermark to generated clips, allowing AI-made video to be identified. That kind of provenance technology will likely become less of a side note and more of a basic expectation as synthetic media moves from novelty to everyday office life.

The timing is telling. For years, workplace communication has been moving away from the long memo and toward a mix of short videos, recorded presentations, visual explainers and asynchronous updates. Remote and hybrid work accelerated the trend, but the underlying need is broader: people want information that is quicker to absorb and easier to share. Video is often better than text for demonstrating a product, explaining a process or delivering something with a human tone.

The catch has been production. Writing a few paragraphs in a document is low-pressure; producing a video can feel like a small event. Tools such as Vids are betting that generative AI can shrink that gap. Instead of asking employees to become editors, designers and presenters, the software takes on more of the technical labor.

There is an upside to that democratization. Teams with fewer resources can make better explainers. Small businesses can give internal communications more polish. Subject-matter experts who are comfortable writing but camera-shy can still contribute video. A rough concept no longer needs to survive a complicated handoff from writer to designer to editor before it can be shared.

Still, lower barriers can create a new problem: more video than anyone has time to watch. If every update becomes a generated clip, the workplace could trade overloaded inboxes for overloaded feeds. The useful test will not be whether AI can make a video quickly. It clearly can. The test is whether the video says something worth stopping for.

That may be where the best use of these tools lies. The strongest AI-generated workplace videos will probably not be the flashiest ones. They will be the ones that make a complicated idea clearer, give a scattered team a better sense of direction or help a real person communicate when the logistics of being on camera get in the way.

Google says Gemini Omni and personal avatars are rolling out in Vids to Google AI Pro and Ultra subscribers as well as Google Workspace business customers. Personal avatars are currently limited to eligible users aged 18 and older in certain regions.

For now, the announcement is less about replacing human video than removing the obstacles that keep people from making it. The result could be a workplace where more messages arrive with a face, a voice and a bit of movement — even when the person behind them never had to find the right lighting.