Gemini AI Now Books Rides and Orders Food via Mobile Apps on Samsung and Pixel Phones

Google’s Gemini AI has taken a transformative step into real-world task automation, now capable of booking Uber rides and ordering DoorDash meals directly through smartphone apps — without user input beyond a verbal request. The feature, debuting on the Samsung Galaxy S26 and Google Pixel 10, represents the first time a consumer-grade AI assistant has been granted agency to perform actions within third-party applications, signaling a new era in mobile AI integration.

According to AndroidPolice, the new functionality leverages on-device AI enhancements that allow Gemini to interpret user intent and execute complex workflows across app interfaces. This isn’t mere voice command automation; Gemini actively navigates the UI of apps like Uber and DoorDash, selects options, confirms orders, and even handles payment authentication using biometric verification tied to the user’s device. The system uses contextual understanding to adapt to app layouts, making it resilient to minor interface changes — a significant advancement over previous AI assistants that required rigid, pre-programmed commands.

The Verge highlights this as Gemini’s first true "agentic" capability, meaning it doesn’t just respond to queries but initiates and completes tasks independently. In a live demonstration, a user simply said, "Order me a pepperoni pizza and book a ride home," and Gemini opened DoorDash, filtered for highly-rated restaurants nearby, selected a meal based on past preferences, applied a discount code, confirmed delivery details, and then switched to Uber to schedule a pickup 15 minutes after the estimated delivery time. All of this occurred without the user touching the screen.

ChromeUnboxed notes that this functionality is powered by a new AI layer called "App Interaction Engine," which maps UI elements in real time using computer vision and natural language understanding. Unlike earlier automation tools that required screen scraping or API access, Gemini operates as a human would — interpreting buttons, text fields, and menus visually. This approach ensures compatibility with any app, even those without developer APIs, vastly expanding its potential use cases.

Privacy and security remain central to the design. Google emphasizes that all processing occurs on-device for sensitive tasks like payment and location sharing. Users must explicitly enable agentic mode and can review and revoke each action before confirmation. A transparent audit trail logs every automated task, accessible via the Gemini settings panel.

Industry analysts suggest this move positions Google to compete directly with Apple’s Siri and Amazon’s Alexa, both of which have struggled with cross-app automation. By embedding AI directly into the Android OS and partnering with Samsung, Google is creating a seamless ecosystem where AI doesn’t just answer questions — it acts.

While currently limited to the Galaxy S26 and Pixel 10, Google plans to expand the feature to other Android 15+ devices later this year. Developers are also being invited to optimize their apps for Gemini’s interaction patterns, potentially leading to a new standard in AI-native app design.

For consumers, the implications are profound: daily routines — from meal planning to commute coordination — may soon require minimal human intervention. As Gemini evolves, its ability to learn individual preferences and anticipate needs could redefine the relationship between users and their devices. But with greater autonomy comes greater responsibility. As this technology becomes ubiquitous, questions around oversight, consent, and unintended actions will demand careful attention from regulators and designers alike.