Google rolls out Gemini Tasks mode in testing with Computer Use and Live voice
Trusted testers gain access to desktop Tasks alongside granular app controls and real-time voice integration.
What to know
- Google has moved Gemini's Computer Use functionality into a public testing phase via a new Tasks mode available to trusted testers.
- Users can now set granular permissions controlling which applications Gemini can operate on their machines.
- Integration with Gemini Live enables voice-driven task execution, allowing users to instruct Gemini to perform desktop tasks in real-time conversation.
- Full production rollout could take weeks or longer; the Windows desktop app itself launched publicly just earlier in September.
Google AI company
How it unfolded 1 development · click the chart to see its coverage articles
-
1
Google ships Gemini Tasks mode to trusted testers with granular app controls
The desktop app update introduces a toggle between Chat and Tasks modes, with access beginning for select testers. A new settings menu lets users configure which apps Gemini can and cannot operate. The Tasks interface also includes an entry point to Gemini Live, enabling voice-driven task execution.
“Tasks, like regular Chat, includes an entry point to Gemini Live. This suggests users could eventually operate Computer Use tasks through real-time voice, asking Gemini to work with their laptop while maintaining a live conversation.”
— TestingCatalog AI News · source -
first by testingcatalog.com, 1d ago · also TestingCatalog AI News
-
-
background
Gemini project creation from folders enters trusted testing — A previously spotted feature allowing users to initiate Gemini projects directly from a folder becomes more prominent and moves into the testing phase.
-
background
Google releases Gemini 3.8 Live with asynchronous function calling — The new real-time audio model gains the ability to make function calls and carry out tasks during dialogue, creating infrastructure for voice-driven task automation.