What is a cloud computer and why does it matter?

Your files, your knowledge, and your AI, accessible from every device, stored in the cloud, with infinite capacity and zero local storage consumed. That's what a cloud computer is.

The term "cloud computer" has historically meant a virtual desktop: a Windows or Linux machine running on a server somewhere, accessed through your browser. That model is about remote access to a generic computing environment. It solves a corporate IT problem (managing desktops at scale) more than it solves a personal productivity problem.

There's a more interesting version of the concept emerging in 2026: a cloud computer as a personal knowledge environment. Not a virtual Windows desktop but a system where your files, your notes, your research, your client work, and your AI assistant live in the cloud, mount on your local device as a drive, consume zero local storage, and are accessible from every device you own. The "computer" isn't a virtual machine. It's a virtual knowledge workspace with AI built in.

How it works

Fabric's virtual drive is the clearest implementation of this concept. It mounts as a folder on your Mac or Windows computer. To your operating system, it looks and behaves like a local drive. You can drag files in. You can open files from it in any application. You can browse it in Finder or Explorer.

The difference from a local drive: the files are stored in the cloud (your own cloud storage with bring-your-own-storage), streamed on demand, and consume zero local disk space. A library with 50GB of accumulated knowledge takes up no more local space than an empty folder. Recently accessed files are cached locally for speed. Everything else streams when you need it.

The difference from Dropbox or Google Drive: every file in the virtual drive is understood by the system. Semantic search finds files by meaning, not filename. The AI assistant can read, summarise, and answer questions about any file in the drive. Smart organisation categorises content automatically. The files aren't just stored in the cloud. They're understood by AI in the cloud.

The difference from a traditional cloud desktop (AWS WorkSpaces, Azure Virtual Desktop): you're not running a virtual operating system. You're running your own operating system with a cloud-backed knowledge layer mounted alongside your local files. Your apps are local. Your knowledge is in the cloud. The two work together seamlessly because the virtual drive presents cloud content as if it were local.

Why it matters

Infinite capacity without local constraints. Your laptop has 256GB or 512GB of storage. Your knowledge library can grow to terabytes without consuming any of it. The constraint shifts from "how much disk space do I have?" to "how much knowledge have I accumulated?"

Every device, same library. Open your laptop: your full library is there. Open the web app on a tablet: same library. Open the mobile app on your phone: same library. The knowledge isn't on a device. It's in the cloud, accessible from every device. No sync delays. No "I left that file on my other computer."

AI that scales with your library. A local AI assistant is constrained by local compute and local storage. A cloud-based AI assistant scales with the cloud: the semantic search handles millions of items, the AI assistant processes documents of any size, and the agents run on schedule without your computer being on.

Data ownership without data custody. With bring-your-own-storage, the cloud storage is yours (S3, R2). The virtual drive reads from and writes to your bucket. If you stop using the tool, your data is in your bucket in standard formats. You get the convenience of cloud without surrendering custody of your data.

Compatibility with every application. Because the virtual drive presents as a standard filesystem, every application that can open a file from a folder can open a file from the cloud computer. Open a PDF in Preview. Open a spreadsheet in Excel. Open an image in Photoshop. The application doesn't know the file is in the cloud because the operating system handles the abstraction.

The broader trend

The cloud computer concept is part of a larger shift: from tools that store your work to tools that understand it. Dropbox stores files. Google Drive stores files. iCloud stores files. None of them understand what the files contain, how they relate to each other, or what you might be looking for when you search.

The cloud computer adds the intelligence layer: semantic search across all content, AI that can reason about your files, automatic organisation, and connections to the other tools where your knowledge lives (email, Slack, meetings, CRM). The "computer" isn't defined by processing power. It's defined by the intelligence it applies to your accumulated knowledge.

For individuals, this means a personal knowledge environment that's available everywhere, understood by AI, and growing smarter with every piece of content you add. For teams, it means a shared knowledge layer that every team member can access from any device, with AI that can answer questions about the company's collective intelligence. The cloud computer is the infrastructure that makes both possible.

Frequently asked questions

How is this different from iCloud Drive or OneDrive? iCloud Drive and OneDrive provide cloud storage with sync. The cloud computer adds AI: semantic search, an AI assistant, smart organisation, self-writing documentation, and agents. Your files are stored and understood, not just stored.

Does the virtual drive work with all applications? Yes. Any application that can open files from a folder can open files from the virtual drive. The operating system treats it as a standard drive.

How much local storage does it use? Minimal. Recently accessed files are cached locally for speed. The bulk of your library streams on demand. A library with tens of gigabytes of content typically uses a few hundred megabytes of local cache.

What if I lose internet connection? Cached files remain accessible. New files can't be added to the virtual drive without a connection. Other capture methods (notes, voice memos) work locally and sync when connectivity returns.

Is this secure? Bring-your-own-storage means data lives in your cloud storage, encrypted with your keys. The virtual drive streams content through encrypted connections. The security is your cloud provider's security (S3, R2), which you control.

What operating systems are supported? The virtual drive works on macOS and Windows. The web app and mobile apps (iOS, Android) provide access from every other platform.

Can teams share a cloud computer? A shared workspace gives the team a shared knowledge layer accessible from every team member's device. Personal content stays private. Shared content is accessible to the team. The same architecture works for individuals and teams.

How does pricing work? Fabric's subscription covers the software and AI. Your cloud storage (S3, R2) costs are separate and typically negligible for personal use (a few dollars per month for most libraries). The total cost is comparable to a premium cloud storage subscription but includes AI, search, agents, and self-writing docs.

Related reading: Why getting things into your second brain is so hard, Your second brain should be compatible with everything, Where does your knowledge live?. Related pages: Sync and backup, Private and secure, Search, Desktop app.