Wednesday, October 21, 2026

AIINEWS

Global intelligence on artificial intelligence and computing

Trusted by Google to grant certification, the Chinese manufacturer installed a backdoor into 1.3 million TV Boxes to sell access rights to hackers.

Source:GenK
05 November, 2026•4 min read
Trusted by Google to grant certification, the Chinese manufacturer installed a backdoor into 1.3 million TV Boxes to sell access rights to hackers.

Executive Briefing

Synthesized and verified with gemini-2.0-flash-thinking-exp

“**Google said that these malicious codes were not errors in the software development process but wer”

Key Takeaways

  • 1.**Google said that these malicious codes were not errors in the software development process but were intentionally installed by the manufacturer on these devices.** Instead of having to find a way to infiltrate from the outside, this time cybercriminals have control over millions of devices right from the factory.
  • 2.Notably, the manufacturer behind these TV Boxes silently installed backdoors into the products, then sold access rights to criminal groups.
  • 3.Devices that have been certified by Google for Android TV thus become a network serving fraud and attacks on the Internet.

Google said that these malicious codes were not errors in the software development process but were intentionally installed by the manufacturer on these devices.

Instead of having to find a way to infiltrate from the outside, this time cybercriminals have control over millions of devices right from the factory. Notably, the manufacturer behind these TV Boxes silently installed backdoors into the products, then sold access rights to criminal groups. Devices that have been certified by Google for Android TV thus become a network serving fraud and attacks on the Internet.

The company discovered is Videostrong Technology, an electronics manufacturer headquartered in Shenzhen, China. This used to be a Google certified partner in the Android TV ecosystem. However, Google's investigation discovered that Videostrong actively installed malware into the firmware of more than 1.3 million devices, mostly TV Boxes running Android TV and Google TV.

Unlike regular malware infections, users do not need to download a suspicious application or access a dangerous link. The backdoor has been integrated right into the production process, before the device reaches the customer. Videostrong then sells access to these devices to cybercriminal groups, and also helps download third-party malicious code modules to the devices.

Once controlled, TV Box can be turned into a tool for many illegal activities. Attackers can install additional applications without the user's knowledge, execute commands remotely, collect device and account information, and even survey the home's internal network. Some functions also allow taking screenshots, recording keystrokes and disabling the Google Play Protect application verification mechanism.

More worryingly, these TV Boxes can be turned into a link in the residential proxy network. At that time, the user's home Internet IP address is used as a traffic transfer point for cybercriminal activities. The device owner may not know that the family's Internet connection is being used for illegal purposes.

Google said this was not an individual employee's behavior or a mistake in the software development process. The malicious code system is maintained for many years, comes with its own server infrastructure and constantly changing configurations to avoid detection. Since late 2024, the malware has evolved into a three-stage architecture, allowing for flexible deployment of various modules.

To bypass the certification process, Videostrong uses a notable trick: malicious code is pre-installed in the firmware but only downloads or activates dangerous components after the user first turns on the device. Thanks to this, the product can pass compatibility and safety tests before being certified. This vulnerability also caused Google to admit that its device evaluation process was not strict enough.

Another problem is that the backdoor is difficult to remove using normal means. The malicious component is placed in the system partition and operates with elevated Android permissions. Even if the user removes it as an app, the malicious code can reinstall itself after the device restarts.

Faced with the scale of the incident, Google ended its partnership with Videostrong, revoked its certification partner status, and banned the company and its affiliates from manufacturing Android devices pre-installed with Google apps or providing software to certified manufacturers. Google also took legal measures to disable the domains and servers that control the XCPE botnet.

However, taking down the control server does not mean the malicious code has been removed from the entire device. This measure essentially cuts off communication between the attacker and the infected machines, thereby preventing them from continuing to receive commands remotely. Google said the campaign also helped block communications of at least 10 million other devices with attackers, belonging to other malware families.

The incident exposes a worrying paradox in the smart device ecosystem: users can buy a product with official certification, but that doesn't mean the manufacturer behind it is always trustworthy. In Videostrong's case, the danger does not come from an app that the user accidentally installed, but from the software itself that was included on the device by the manufacturer in the first place.

Attribution & Original Source

Synthesized and verified from GenK. All copyrights and intellectual property belong to the original publication.

Visit Original Source
Share this intelligence report:

Related Dispatches

Oracle wants to pull stalled AI agents out of demo mode and into production
SiliconANGLE12 minutes ago

Oracle wants to pull stalled AI agents out of demo mode and into production

### Oracle wants to pull stalled AI agents out of demo mode and into production Companies want AI agents faster than they can build the AI architecture to run them. Most can put together a convincing demo. Far fewer can move agents into production without exposing data, hurting uptime or blowing their budgets. That gap is pushing database providers to rethink what a data platform for agents has to deliver. Nearly 70% of organizations plan...

4 min read
OpenAI says its new cybersecurity model is better at not hacking other companies
The Verge13 minutes ago

OpenAI says its new cybersecurity model is better at not hacking other companies

OpenAI is launching GPT-6 Cyber, the latest version of its cybersecurity AI model, which it says is better at finding vulnerabilities and less likely to go rogue and hack other platforms in the process. OpenAI says that during testing, GPT-6 Cyber went out of bounds less than 1 percent of the time, down from around 5 percent with GPT-5.6 Cyber. # OpenAI says its new cybersecurity model is better at not hacking other companies Classified as...

3 min read
Oracle’s multicloud approach: Put the database wherever AI pipelines already run
SiliconANGLEabout 1 hour ago

Oracle’s multicloud approach: Put the database wherever AI pipelines already run

### Oracle’s multicloud approach: Put the database wherever AI pipelines already run Enterprises racing to put AI agents into production are learning that real-time data pipelines matter as much as the models. An agent working from a stale subset of corporate records delivers far less than one wired into live, mission-critical data. That shift is changing how companies decide where their databases run. Oracle Corp. has pushed its database...

3 min read7