
Gemini Spark launches in the US for Google AI Ultra subscribers. It is a 24 hour personal agent that automates tasks via a new side panel tab on web, Android, and iOS. The feature integrates with Google Workspace, Connected Apps, signed-in websites, location, and personal intelligence. Gemini Spark operates through three components: a Task defines the goal, a Schedule sets when it runs, and a Skill provides reusable instructions and tools. Users can assign Skills using @ or / syntax and specify which Skill to use for a Task or Schedule. The Beta status indicates limited availability and ongoing refinement.
Tap to vote and see what everyone thinks.
Gemini's new AI agent is about as good as Google's demo
Summary by ByteBrief