Realtime AI News
Developers are using AI agents to strip ads and bloatware from smart TVs
Developers and vibe coders are pointing agents like Claude Code at their televisions to remove system ads, disable viewing tracking and rewrite the launcher, with public GitHub projects such as dumbtv and cleanyourtv leading the charge. The work needs no root access, and thousands of owners have already tried it on sets from Xiaomi, TCL, Sony, LG and Fire TV.

A loose wave of developers and "vibe coders" is using AI coding agents to take over smart TVs — stripping out advertising, disabling viewing tracking, replacing sluggish launchers and redesigning home screens. TechCrunch reported on Oct. 19 that what began as scattered experiments has turned into shared tooling that thousands of people have tried on their own sets.
The most visible project is called dumbtv. Its GitHub page states the mission plainly: a TV should do what you paid for — your apps, your HDMI inputs and nothing else, with no ads from the set itself, no watch tracking and no updates you did not ask for. A second project, cleanyourtv, promises to strip bloatware from any Android TV or Google TV over Wi-Fi and swap the launcher for one that resembles Apple TV.
cleanyourtv's creator, Gathin T., said the idea began one afternoon when he got tired of the ads on his parents' set. He handed the model number to Claude Code; the agent found the television on Wi-Fi, removed 25 Xiaomi packages and wrote a new launcher. After he published the prompt on GitHub and posted before-and-after clips, a few thousand people did the same over the next two days, across Xiaomi, TCL, Sony, Fire TV, LG and even a projector or two.
What separates this from older TV-hacking guides is the missing technical bar. The solutions do not require root access, and users can return a set to factory defaults if they change their minds. dumbtv's developer says the approach has been tested end-to-end on a Fire TV Edition television and on Android/Google TV hardware from Sony, TCL, Philips, Hisense, Xiaomi and Shield.
The article also documents people improvising. Ryan Doyle said on X that he simply sent Claude Code his TV's product listing and asked whether it could hack the device; the agent walked him through removing ads and bloatware, speeding up a lagging interface and customising the screensaver. Another user, Mihir Gupta, wrote that one afternoon with Claude Code disabled 25 preinstalled apps, replaced a 192 MB stock launcher with an 83 MB custom one and added rotating wallpapers.
The demand behind these experiments is structural. Modern smart TVs ship with Automatic Content Recognition, which identifies what is playing using screenshots, audio clips or other forms of digital fingerprinting. If the de-bloating habit spreads to mainstream consumers, it could chip away at a smart TV advertising business that profits from selling viewing data to advertisers or using it to target ads on the set's own interface.
The risks are as concrete as the appeal. Every one of these projects is a work in progress, with users testing across platforms and filing fixes. Some found that Claude would not respond to a plain request to hack my TV without more elaborate prompts, while others warned that an agent can disable important packages by mistake. One person said they bricked an Android TV with Claude, and another got stuck halfway through a Fire TV.
For people who would rather not hand an agent the keys, Android TV already has established home-screen replacements such as ATV Launcher, Projectivity Launcher and FLauncher. The difference now is that a general-purpose agent can find a set on the network, work out which packages to remove and generate a new interface in a single session — which is exactly the shift the ad business should be watching.
Sources
Why it matters
AI agents have collapsed a niche modding chore into a few prompts any owner can try, putting pressure on the ACR-driven ad model that subsidises cheap smart TVs. The same ease of use cuts both ways: users have already reported bricked sets and broken packages after handing control to an agent.
Nearby Updates
All10/19, 23:54
Sumsub launches a KYA module aimed at stopping unauthorized AI agent actions
Sumsub has launched a KYA, or Know Your Agent, module intended to prevent AI agents from taking unauthorized actions, according to FF News. The release extends identity governance from human users to automated agents, a layer that is becoming a competitive front for identity vendors as companies hand more workflows to software that can act on its own.
10/19, 23:37
Nvidia ships open-weight Nemotron 3.5 Super VL for enterprise agents and Bolt models for local PCs
Nvidia released open-weight Nemotron 3.5 Super VL, a 120-billion-parameter model for enterprise agents, alongside two lightweight Bolt models built to run locally on new AI PCs and workstations. Nvidia claims Super VL beats Anthropic and OpenAI models on specific post-trained agentic tasks, and names Skydance, EY, Ramp, CrowdStrike, Perplexity and Salesforce as users.
10/20, 01:04
Tencent's Hy-Memory Claims Open-Source Benchmark Lead Over Claude Opus 5
Tencent has released Hy-Memory, an open-source model it says leads the open-source field on benchmarks it defined itself, with Claude Opus 5 as the comparison point. Because the results are vendor-reported, the claimed margin still depends on third-party reproduction and real developer adoption.
10/19, 23:00
NVIDIA and the EBU widen their sovereign AI framework to cover agentic security in media
At a Berlin creative-industry event on October 19, NVIDIA said the European Broadcasting Union is expanding the sovereign AI standards framework the two introduced last year, adding sovereign cloud providers, open model developers and integrators to govern how broadcasters deploy agents. Adobe also showed Frame.io video search accelerated by NVIDIA GPUs, claiming roughly 5x faster embedding generation, with German producer UFA already using it across 15 years of footage.