People Are Using AI to Turn Their Smart TVs Dumb Again

Developers are now using AI agents like Claude Code to strip ads, tracking and bloatware off their smart TVs — and thousands of people have copied their prompts in a matter of days.

A wave of "vibe-coding your TV" projects has surfaced recently, with creators sharing solutions for removing sponsored ads, speeding up lagging app launchers, redesigning home screens and disabling tracking features. The movement taps into a frustration that's been building for years: people pay hundreds for a television, then find it serving them ads and logging what they watch.

Two projects in particular are getting attention. One, called "dumbtv," states its mission plainly on its GitHub page: "Your TV should do what you paid for: your apps, your HDMI inputs, and nothing else. No ads from the TV itself, no watch tracking, no update you did not ask for." Another, "cleanyourtv," promises to strip bloatware from any Android TV or Google TV over Wi-Fi and replace the stock launcher with one resembling Apple TV's.

Takeaway: the pitch isn't new features — it's subtraction, and the tool doing the subtracting is an AI coding agent.

What started it?

The creator of cleanyourtv, Gathin T., says the project began at his parents' house. He got tired of seeing ads on their TV set and decided to do something about it.

"I had Claude Code clean it up over Wi-Fi and build a simple home screen with just their apps on it. I put the prompt on GitHub, tweeted the before and after, and a few thousand people did the same to their own TVs over the next two days. Xiaomi, TCL, Sony, Fire TV, LG, a projector or two," Gathin wrote on a donations page linked from the project's homepage.

A few thousand installations in 48 hours is small change in the consumer electronics market, but it signals demand. And the barrier to entry is remarkably low: the solution doesn't require root access to the TV, and users can return to the factory default state whenever they want.

The dumbtv project has been tested end-to-end on a Fire TV Edition television and on Android/Google TV platforms from Sony, TCL, Philips, Hisense, Xiaomi and Shield. To use it, the developer suggests pointing your preferred AI agent at the dumbtv repository on GitHub.

Takeaway: one prompt, shared publicly, was enough for thousands of owners to replicate the fix on major TV brands within two days.

Can you just ask an AI to "hack" your TV?

Increasingly, yes. In a viral X post, Ryan Doyle explained that he sent Claude Code the product listing for his TV and asked "can you hack this?" The agent then talked him through the process. Doyle claims it let him remove ads and bloatware, speed up the lagging interface and customize his screensaver — and he demonstrates the responsiveness of the cleaned-up interface in a follow-up video.

The posts inspired others to try it. X user Mihir Gupta documented his results in concrete terms: "One afternoon with Claude Code: 25 bloat apps disabled, the 192 MB stock launcher replaced with a custom 83 MB one, plus rotating wallpapers."

That's the mechanism at work here. Hacking consumer electronics for more control isn't new — tinkerers have done it for decades. What's changed is accessibility. You no longer need to read forum threads and write scripts yourself; an AI agent reads the documentation, writes the commands and walks you through them in plain conversation.

Takeaway: the technical skill that used to gate this kind of modification has been outsourced to the model.

Why does this matter to the smart TV industry?

Money and data. Modern smart TVs ship with a technology called Automatic Content Recognition (ACR), which identifies what's playing on the screen using screenshots, audio clips or other forms of digital fingerprinting. That viewing data feeds an ad ecosystem: the smart TV ad industry profits from selling viewing data to advertisers or using it to display personalized ads on the TVs' own interfaces.

If these "dumb TV" hacks spread beyond hobbyists to everyday consumers, they have the potential to disrupt that business model. A TV that runs only your apps and your HDMI inputs generates nothing for the data pipeline. Manufacturers and ad-tech firms have built revenue expectations on top of interfaces users are now learning to dismantle with a chat window.

Takeaway: every cleaned TV is a small hole in the automatic-content-recognition economy that smart TV advertising depends on.

What can go wrong?

Plenty. All these solutions are works in progress, and users are currently testing them across different TV platforms, suggesting updates and fixes as they go.

The results are mixed. Some people found that Claude wouldn't respond to a simple "hack my TV" command, forcing them to write more complex prompts. Others warned that AI agents can make mistakes — potentially disabling important system packages or breaking functionality along the way. One person said they bricked their Android TV using Claude. Another got stuck halfway through hacking their Fire TV.

There's also a safer, older route for the cautious. Customization apps for Android-based smart TVs already exist, including home screen replacement launchers like ATV Launcher, Projectivity Launcher and FLauncher. They require more manual setup, but they carry less risk of an AI agent deciding to remove something your TV actually needs.

Takeaway: an AI that can strip a launcher can also strip the wrong package, and at least one user has ended up with a dead TV.

Where does this go next?

The immediate trajectory depends on whether these tools stay niche or get packaged for mainstream use. The ingredients for wider adoption are already visible: prompts that work across brands from Xiaomi to Sony, no root access required, reversibility to factory defaults, and results measured in gigabytes of bloat removed. What stands in the way is reliability — bricked sets and half-finished hacks don't make for confident recommendations to less technical users.

For TV makers and the advertisers who buy their viewing data, the signal is harder to ignore: the same AI coding agents now writing software for millions of developers can also be pointed at a consumer device and told to take out the parts that make money.

By Sophie Lindqvist