Picture the closet. Behind the winter coats and the box of cables you’ll never identify, there’s a laptop. It boots in four minutes. The fan sounds like a small aircraft. You keep it because throwing away a computer feels wasteful, and buying a new one feels expensive. So it sits there, aging quietly.
Mukesh Ambani would like a word with that laptop.
Reliance Jio, the telecom arm of Ambani’s conglomerate, is rolling out a service that turns old computers into what it’s calling AI-ready PCs. The mechanism is cloud services, and the price is roughly $11 for two months. The stated goal is to stretch the useful life of hardware people already own instead of pushing them toward a new machine. TechCrunch’s Jagmeet Singh reported the details, and the product is called JioPC.
I want to unpack what’s actually happening here, because “AI-ready PC” sounds like marketing until you understand where the intelligence lives.
Your computer becomes a window
Here’s the mental model that helps. Most of us assume a computer’s abilities live inside the box on our desk. Faster chip, better performance. More memory, more browser tabs. That intuition has been correct for about forty years.
Cloud computing flips it. The actual processing happens in a data center somewhere, on hardware you’ll never see. Your machine’s only job is to show you the picture and send back your clicks and keystrokes. It’s less like a brain and more like a window into a brain that lives elsewhere.
Once you frame it that way, the aging laptop problem dissolves. A slow processor doesn’t matter much if it isn’t doing the heavy work. What matters is the screen, the keyboard, and a decent internet connection. Your creaky machine becomes a perfectly good window.
This isn’t a new idea. Anyone who used a computer lab in the nineties met its ancestor. What’s changed is that AI workloads are unusually well suited to it, because they’re expensive to run locally and cheap to reach over a network.
Why the price tag is the real story
Eleven dollars over two months is not a rounding error in the AI industry, it’s a statement about who the customer is. Reliance is treating access to AI computing the way telecom companies treat mobile data: something you subscribe to in small increments, not something you buy in one painful transaction.
Reliance has publicly committed to a large AI expansion, with Forbes noting 10 trillion rupees, about $110 billion, earmarked for those plans. The reporting frames Jio’s ambition as becoming India’s default gateway to AI computing. A cheap subscription that runs on hardware people already have is a reasonable way to pursue that. You skip the biggest barrier, which is asking someone to buy a new laptop first.
What this means if you care about AI agents
This is the part that interests me most, and it’s why I think readers here should pay attention to an Indian telecom product even if they’ll never subscribe.
AI agents, the kind that take a goal and work through steps on your behalf, are awkward to run on personal hardware. They’re hungry. They need models, they need to keep running while they work, and they often need to talk to other services. A laptop from 2016 is not where you want that happening.
But an agent doesn’t especially care where it lives. It needs compute and a connection, not proximity to you. So the cloud model that rescues an old laptop is the same model that makes agents practical for people who don’t own expensive machines. If your computer is already a window, adding an agent behind the glass is a straightforward extension rather than a hardware upgrade.
The direction that points is worth sitting with. For a lot of people, their first real experience of an AI agent won’t arrive as software they install. It’ll arrive as something already waiting inside a subscription they pay monthly.
The questions I’d ask before signing up
I don’t want to oversell this. A few things follow directly from how the model works, and they’re worth understanding before you get excited.
- No connection means no computer. When the processing is remote, a dropped network isn’t an inconvenience, it’s the whole machine going dark.
- Your work lives somewhere else. That’s a convenience and a dependency at the same time, and the terms of that arrangement matter.
- Subscriptions are not one-time purchases. Eleven dollars is small. Eleven dollars forever is a different calculation, and it’s the calculation the provider is making too.
None of that makes the idea bad. It makes it a trade, and trades are easier to evaluate when you can see both sides.
What I like about this move is how unglamorous it is. Most AI news arrives wrapped in claims about intelligence and the future. This one is about a slow laptop in a closet and whether it can be useful again for the price of two coffees. That’s a smaller question with a much bigger audience.
🕒 Published: