Google rebrands Android AI as Gemini Intelligence, adds agentic task automation and vibe-coded widgets

Google is bundling its most advanced on-device AI capabilities under a new name: Gemini Intelligence. According to The Verge, the company's director of Android experiences, Ben Greenwood, said the branding brings the best of Gemini to the most advanced Android devices. The move appears designed to differentiate premium AI features from the baseline Gemini assistant, reserving them for flagship phones like the Galaxy S26 series.

TechCrunch reports that the suite includes Gboard-based dictation, form-filling capabilities, and the ability for AI to complete tasks across multiple apps. The features build on agentic capabilities Google first demonstrated at Samsung's Galaxy S26 launch earlier this year, where Gemini could order food or book a ride. Now the assistant can handle more complex chains: finding a spin class, locating the syllabus in Gmail, and searching for related books, all in sequence.

Bloomberg frames the timing as strategic, noting the announcement comes weeks before Apple is expected to unveil major Siri upgrades. The competitive pressure is unmistakable. Google isn't just adding features; it's packaging them into a premium tier that gives its hardware partners a selling point against whatever Apple has planned.

The core promise of Gemini Intelligence is that your phone does things for you, not just answers questions. The Verge describes the vision as Gemini showing up in more places: Chrome on Android, autofill suggestions, and deeply integrated across apps, if the user opts in. The assistant can browse the web, fill out forms, and complete multi-step workflows that span different applications.

TechCrunch details the progression from earlier agentic demos. At the Galaxy S26 launch, Gemini handled straightforward transactions like ordering food or booking a ride. The new capabilities add complexity: the assistant can now chain together tasks that require context from multiple sources. Booking a front-row bike for a spin class involves checking availability, while cross-referencing a class syllabus in Gmail with book searches requires understanding relationships between documents.

Bloomberg emphasizes that these features are part of Android 17, signaling deep OS-level integration rather than a bolt-on app experience. The agentic layer sits close to the system, which is what enables it to move between apps, read screens, and take actions on the user's behalf. This isn't a chatbot floating above the OS; it's woven into the fabric of how the phone operates.

One of the more surprising announcements is Create My Widget, which TechCrunch describes as a feature that lets users vibe code their own custom Android widgets using natural language. Launching first on the latest Samsung Galaxy and Google Pixel phones this summer, the tool lets anyone describe what they want and get a functional, resizable widget on their home screen.

TechCrunch offers concrete examples: a user could ask for three high-protein meal prep recipes every week, and Gemini builds a custom dashboard pulling from the web and Google apps like Gmail and Calendar. A cyclist who only cares about wind speed and rain can create a weather widget that surfaces exactly those stats and nothing else. The system connects to multiple data sources to build a single, personalized view.

The Google AI Blog frames this as part of a broader full-stack vibe coding experience in Google AI Studio, suggesting the widget feature is a consumer-facing expression of a developer trend Google has been cultivating. The term vibe coding, popularized in developer circles for describing AI-generated code through natural language prompts, is now being productized for everyday Android users who would never open a code editor.

While consumers get vibe-coded widgets, developers are getting their own agentic tools. TechCrunch reports that Android Studio is adding a Journeys feature and Agent Mode, bringing AI assistance directly into the development environment. The details are sparse, but the direction is clear: Google wants AI to help build the apps that AI will then help users navigate.

The symmetry is intentional. On one side, Gemini Intelligence automates tasks for phone owners. On the other, agentic AI in Android Studio automates tasks for the people building those experiences. Google is creating a feedback loop where AI assisted development produces apps designed for AI assisted usage. The Google AI Blog's mention of a full-stack vibe coding experience in AI Studio reinforces this two-sided strategy.

This developer push matters because it creates an ecosystem moat. If Android becomes the easiest platform for building AI native apps, developers have a strong incentive to target it first. The agentic features in Android Studio lower the barrier to entry while the Gemini Intelligence features on devices create demand for apps that can be controlled by AI.

Beyond software, Google announced Googlebooks, a new line of AI-first laptops built with Gemini at their core. TechCrunch reports that Google is working with partners including Acer, Asus, and Dell, suggesting a broad hardware strategy rather than a single first-party device. The laptops are positioned as vehicles for Gemini Intelligence, extending the AI assistant beyond phones.

The Verge notes that Gemini is also coming to Chrome on Android, which bridges the mobile and desktop experiences. A user could start a task on their phone and continue on a Googlebook, with Gemini maintaining context across devices. This cross-device continuity is something Apple has emphasized with its ecosystem, and Google appears to be building a parallel AI-powered version.

Bloomberg's framing of the announcements as a preemptive strike against Apple's upcoming Siri revamp takes on added weight with the hardware dimension. Google isn't just upgrading software; it's launching devices designed specifically for an AI-first experience, potentially setting a new baseline for what users expect from laptops.

Bloomberg places the announcements in direct competitive context: Google is showcasing its AI progress weeks before Apple is expected to announce major Siri upgrades. The timing is not coincidental. Google wants to define the conversation around on-device AI before Apple has a chance to reframe it.

The Verge captures the sheer volume of the Gemini push, calling it Gemini season and noting that Google just can't help itself when it comes to naming things. The Gemini Intelligence brand joins a crowded field of Google AI product names, but the bundling strategy makes sense: premium features get a premium label, creating clear differentiation for high-end Android phones.

Google's annual I/O developer conference later this month will likely expand on these announcements with more technical detail. The features announced at the Android Show are shipping this summer, starting with Samsung Galaxy and Pixel devices. The question now is whether Apple's response, expected soon, will match Google's pace or take a different approach to on-device intelligence.

Taken together, the announcements paint a picture of a fundamental shift in how smartphones work. Gemini Intelligence isn't about answering questions faster; it's about the phone taking action. Form filling, cross-app task completion, widget creation, and web browsing all point toward a device that anticipates needs and executes on them.

TechCrunch's coverage emphasizes the practical applications: dictation through Gboard, autofill that understands context, and widgets that pull from multiple data sources to create personalized dashboards. The Verge highlights the opt-in nature of the deeper app integration, acknowledging the privacy implications of an AI that can read your emails and move between your apps.

Bloomberg's competitive lens suggests the stakes are high. If Google succeeds in making Gemini Intelligence a reason to choose Android, it puts pressure on Apple to deliver a Siri experience that can match agentic capabilities. The summer launch window gives Google a head start, but Apple's developer conference is right around the corner. The race to build a phone that actually does things for you is officially underway.