Google unveiled plans to integrate agentic artificial intelligence capabilities into its Gemini app on Tuesday, allowing the AI to autonomously handle tasks such as ordering groceries, booking rideshares, and reordering frequently purchased items. The announcement came as part of a broader presentation of new features for Pixel smartphones.
According to a blog post by Michal Levin, Google’s director of product management, Gemini will “work with your apps in the background to complete everyday tasks,” effectively acting as a digital assistant capable of managing routine errands. “Let Gemini handle your busy work,” Levin wrote.
The new functionality, initially announced on February 25th, will launch as a beta feature for users of the Pixel 10, Pixel 10 Pro, and Samsung Galaxy S26 series smartphones, with initial availability limited to the United States and South Korea. The feature is designed to perform multistep tasks although users continue to utilize their phones for other purposes.
Google emphasized safety and privacy considerations in the rollout. The company stated that automations will only initiate with explicit user command, provide ongoing notifications to allow monitoring and interruption, and will be restricted to accessing a limited selection of apps, rather than the entire device. The initial beta will focus on applications within the food, grocery, and rideshare sectors.
DoorDash confirmed its participation in the beta program, announcing on X that users will be able to order groceries and meals through the platform via Gemini. Samsung also highlighted the feature in a press release accompanying the launch of its Galaxy S26 series, noting that tasks can be initiated with a single button press or voice command.
“Once set up, tasks can be completed with a single button press or voice prompt,” Samsung stated. “Galaxy S26 can also handle multistep tasks in the background, streamlining the process on the user’s behalf. For example, with Gemini, booking a taxi is as simple as asking, reviewing the details and tapping confirm.”
Google indicated that this initial release represents only the beginning of Gemini’s agentic capabilities, suggesting further expansion into additional app categories and functionalities in the future.
