From an earlier video
This is material from an earlier video — the interfaces and features of AI services may have changed since. The idea and the approach still hold.
Screen sharing:
🔍 What is screen sharing in Gemini?
It’s a mode where you:
- start sharing your screen in a browser window,
- show the interface of an app, site or document,
- and get live hints, explanations and help from the AI in real time.
🔧 The mode is built on Project Astra — Google’s initiative to create an active, attentive AI that sees, hears and understands its surroundings.
💼 Use cases
🎨 Design and animation
You show the Figma or After Effects interface and ask:
Suggest how to simplify the colour palette and add motion to the animation
The AI sees your screen and proposes concrete actions: change layers, structure, fonts.
🧾 Working with Obsidian
You show your note structure:
How can I visually simplify this map? What can be automated?
Gemini analyses the blocks and suggests a plugin or a Templater command.
🧑💼 Working with documents
You show a Google Doc:
What's unclear in this section? How can I simplify the explanation for schoolkids?
The AI rephrases it, pulls out the key points, suggests examples.
🔐 Privacy and security
- The screen is shared only within your session
- It’s not saved by Google (unless you give separate permission)
- You can pause or stop sharing at any moment
🤖 Voice and gesture integration (Astra)
As part of Project Astra, Google is testing: *Astra isn’t publicly available yet.
- Understanding intonation and speech in real time
- Reactions to gestures, the mouse, movement
- Answers even before you finish your question (intuitive help)
This makes interacting with AI as close as possible to a real mentor.
⚠️ Downsides and limitations
- It doesn’t work in all countries or on all devices yet
- Sometimes it misses the context if there are too many moving elements
- You can’t turn on the camera (screen only)
Keep going?
- 🧠 [[seven_video|Back to the topics of video #7]]
- ⬅️ Back to home