Google killed the Chromebook. Its replacement turns your cursor into an AI agent.

Google killed the Chromebook. It took 15 years, but the company that invented the browser-as-operating-system has concluded that a browser is not enough. At the Android Show on Monday, Google unveiled Googlebook, a new category of premium laptops running Android with Gemini embedded at the operating system level. The devices will ship this autumn from Acer, ASUS, Dell, HP, and Lenovo. ChromeOS is not being updated. It is being replaced.

The rebrand is not cosmetic. Googlebook runs on what Google internally calls Aluminium OS, a version of Android 17 rebuilt as a genuine desktop platform with a custom window manager, native multitasking, and Gemini woven into every interaction. There are no containers, no emulation layers, no compatibility modes. The operating system is Android. The apps are native. The AI is not an assistant sitting in a sidebar. It is the interface.

The cursor

The most revealing feature is called Magic Pointer. Built with Google’s DeepMind team, it turns the laptop cursor into a context-aware AI agent. Wiggle the cursor over a date in an email and Gemini offers to schedule a meeting. Point at two images and it composites them together. Select a paragraph and it summarises, translates, or rewrites. The cursor, the oldest interaction metaphor in personal computing, becomes a direct channel to a large language model that can see your screen and act on what it finds.

Magic Pointer is not a chatbot. It does not require a prompt. It reads the context of whatever the cursor touches and surfaces actions before the user asks for them. The distinction matters because it represents a different theory of how AI should enter personal computing. Apple embeds intelligence into individual applications. Microsoft puts Copilot in a panel beside the workspace. Google is putting Gemini inside the pointing device itself.

The second signature feature is Create your Widget, which lets users describe a custom widget in plain language and Gemini builds it on the spot, pulling data from Gmail, Calendar, web searches, and other Google services into a single personalised dashboard. The widgets are vibe-coded, generated by AI from a natural language description rather than selected from a catalogue. The user does not choose from what exists. The user describes what should exist and the machine builds it.

The unification

Googlebook solves a problem that has plagued Google for a decade. ChromeOS was a web-first operating system that ran Android apps inside a compatibility layer. The experience was functional but compromised. Android apps on Chromebooks ran in containers that could not access the file system natively, could not interact with desktop windows properly, and could not use hardware features the way they could on a phone. The platform had two souls and neither worked as well as it should have.

Aluminium OS eliminates the split. Android apps run natively on the laptop because the laptop runs Android. A feature called Cast my Apps lets users open any application from their Android phone on the Googlebook’s screen without downloading it. Quick Access provides direct access to phone files through the laptop’s file browser, with no manual transfers required. The phone and the laptop share an operating system, an app ecosystem, and an AI layer.

The European Commission is preparing to force Google to give rival AI assistants the same access to Android that Gemini receives, a ruling that could determine whether Googlebook’s deep Gemini integration becomes a competitive moat or a mandated open platform. The DMA decision is expected in July. Googlebook ships in the autumn. The regulatory timeline and the product timeline are on a collision course.

The market

Googlebook enters a laptop market that has been reshaped by two forces. The first is Apple’s MacBook Neo, a 599-dollar laptop running macOS on an A18 Pro chip derived from the iPhone, which brought Apple’s entry price below 600 dollars for the first time. The second is the Snapdragon X Elite, which gave Windows laptops competitive battery life and AI inference capabilities for the first time in years.

Google’s response is to abandon the low end entirely. Googlebook is positioned as a premium product with what Google describes as premium craftsmanship and materials. Every device will feature a Glowbar, an LED strip embedded in the keyboard deck that animates in response to Gemini’s activity. The company is not building a cheap laptop that happens to have AI. It is building an AI device that happens to be a laptop.

The pivot is striking because Chromebooks succeeded precisely by being cheap. Google is deploying Gemini to four million GM vehicles, embedding the same AI into cars, phones, wearables, and now laptops. The pattern is clear. Google does not want Gemini to be a product. It wants Gemini to be the intelligence layer that connects every screen in a user’s life. Googlebook is the desktop-sized piece of that strategy.

The education question

Chromebooks hold more than 60 per cent of the global education laptop market. The platform serves 38 million students in K-12 schools, and 93 per cent of US school districts plan Chromebook purchases this year. The installed base is enormous, the margins are thin, and the switching costs are low.

Google says existing Chromebooks will continue receiving security updates until their stated auto-update expiration dates. Some devices may qualify for an opt-in upgrade to the new platform. But the premium positioning of Googlebook raises an obvious question: what happens to the education market that made Chromebooks ubiquitous? A 200-dollar Chromebook for a fourth-grader and a premium Googlebook for a professional are different products for different buyers. Google has not announced pricing, but the emphasis on premium hardware suggests that the cheapest Googlebook will cost considerably more than the cheapest Chromebook.

Intel is previewing its next generation of AI PC processors at Computex 2026, betting that local AI inference on laptops is the next wave of chip demand. Googlebook’s Gemini integration is cloud-first, but the hardware partnerships with Intel’s competitors in the ARM ecosystem suggest that on-device AI processing will follow. The question of where the intelligence runs, on the device or in the cloud, will determine whether Googlebook works offline and how much it costs to operate at scale.

The strategy

Google’s laptop strategy has always been a distribution strategy for Google services. Chromebooks put Chrome, Gmail, Drive, Docs, and Search in front of hundreds of millions of users, particularly students, at the lowest possible hardware cost. The services generated the revenue. The hardware was the delivery mechanism.

Googlebook extends the same logic but changes the service being distributed. The service is no longer a suite of web applications. It is Gemini. The AI that reads your email, builds your widgets, summarises your documents, and anticipates your next action is the product. The laptop is the surface it runs on. The Android ecosystem, with its 3.6 billion active devices, is the network that feeds it context.

Apple has argued that AI will become as commonplace as word processing, a utility that disappears into the background of everyday computing. Google is making a different argument. Gemini on Googlebook is not in the background. It is the cursor. It is the widget. It is the interface between the user and the machine. Google is not making AI invisible. It is making AI the thing you interact with every time you touch the trackpad.

Apple’s AI rollout has already stumbled in China, where regulatory delays left Cupertino without its most important differentiator in its most competitive market. Google faces similar risks. Gemini’s deep integration into Googlebook means that any market where Gemini is restricted, whether by regulation, data sovereignty requirements, or competitive dynamics, is a market where the laptop’s core value proposition is diminished.

Fifteen years ago, Google bet that the browser was the operating system. That bet built a dominant position in education and a meaningful share of the consumer laptop market. Now Google is betting that AI is the operating system, that the intelligence layer matters more than the application layer, and that the company with the most context about a user’s life, across phone, car, watch, and laptop, will build the most useful computing experience. Googlebook is not a laptop. It is the argument that the device does not matter. The intelligence does.