Gemini task automation rolls out to Galaxy S26 and Pixel 10 with app control for ordering food

Reviewed byNidhi Govil

9 Sources

Share

Google's Gemini can now handle ordering food and booking rides on Galaxy S26 and Pixel 10 devices through a new task automation feature. The AI assistant takes control of apps like Uber Eats and DoorDash, completing multi-step processes based on simple voice commands, though users must confirm final checkout for security.

Gemini Task Automation Arrives on Galaxy S26 and Pixel 10

Google and Samsung have launched a significant update to Gemini that transforms how users interact with their smartphones. The new task automation feature, now live on the Galaxy S26 series and rolling out to Pixel 10 devices as part of the March Feature Drop, allows the AI assistant to take control of specific apps and complete complex workflows on behalf of users

1

5

. This agentic task automation capability marks a shift from traditional voice assistants that merely answer questions to an AI that actively performs tasks within Android apps.

Source: Tom's Guide

Source: Tom's Guide

The feature works through natural language commands, enabling users to make requests like "order a coffee and a croissant" or "get me a ride to the airport" without manually opening apps or navigating through menus

2

. Gemini handles the entire process in a virtual window, adding items to carts, selecting options, and even making contextual decisions—such as specifying that a pastry should be warmed—without explicit user input

1

.

How Screen Automation Feature Works in Practice

When users activate the screen automation feature through conversational commands, Gemini takes over Android app control in the background. Testing on the Galaxy S26 Ultra demonstrated the system's capability to handle ordering food and rideshares with minimal user intervention

4

. For instance, when asked to "order a spicy chicken sandwich from Popeye's on Uber Eats," Gemini initiated the process, added items to the cart, and even skipped unnecessary add-on pages to streamline the workflow

4

.

Source: Tom's Guide

Source: Tom's Guide

The multi-step task completion capability handles complex requests that previously required multiple manual actions. Users receive notifications when Gemini is working on a task and can either watch the process unfold in real-time or continue using their phone for other activities

2

. However, for security purposes, Google does not allow Gemini to press the final "place order" or "pay" button. After Gemini builds the cart and navigates through the ordering process, user confirmation is required to complete the purchase

2

.

Supported Apps and AI Integration for Galaxy Users

At launch, the feature supports a limited but practical selection of apps focused on ordering food and rideshares. Compatible services include Uber, Lyft, Uber Eats, GrubHub, DoorDash, and Starbucks

4

3

. The list of supported apps appears in Gemini's settings and is specific to what's currently installed on each device

4

. Google has opened the door for additional apps to join the ecosystem as the system expands beyond food delivery and rideshare services

2

.

Source: 9to5Google

Source: 9to5Google

The beta feature is currently launching in select markets, including the United States and South Korea, exclusively on Galaxy S26 and Pixel 10 phones

2

. This represents a significant step in AI integration for Galaxy users and the broader Android ecosystem, though availability varies by region and device.

Privacy Concerns and Usage Limitations

While the automation capability offers convenience, it comes with important privacy considerations. The automation process requires Gemini to take screenshots of app activity, potentially exposing visible information. If users have Google services set to "Keep Activity On," these screenshots are reviewed by trained human reviewers

5

. Users concerned about privacy concerns can switch off activity history to prevent human review of their data.

Usage limits also apply based on subscription tier. Free users are restricted to 5 requests per day, while AI Plus subscribers receive 12 requests. AI Pro subscribers get 20 requests, and AI Ultra plan holders can make up to 120 requests

5

. Additionally, early testing revealed that while the feature works as intended, the process can take considerably longer than manual ordering—approximately two and a half minutes versus 30 seconds for traditional app navigation

5

. The primary advantage lies in having Gemini handle the grunt work in the background while users focus on other tasks, rather than pure speed optimization.

As Google continues to refine this capability and expand to more apps, the feature signals a broader shift toward AI assistants that don't just respond to queries but actively automate tasks in apps, potentially reshaping how millions interact with their smartphones daily.

Today's Top Stories

TheOutpost.ai

Your Daily Dose of Curated AI News

Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Spend less time searching for the latest in AI and get straight to action.

© 2026 Triveous Technologies Private Limited
Instagram logo
LinkedIn logo