More than 1.3 million Android devices were infected with malware, and their owners had no idea their devices were being used for fraudulent activities.
On November 2, 2026, Google announced that it had worked with industry partners to dismantle the XCPE botnet, a network of compromised devices used for advertising fraud, residential proxy services, and other malicious activities.
According to the Google Threat Intelligence Group, the malware was deliberately built into devices manufactured by Chinese company Videostrong Technology.
The manufacturer allegedly sold access to criminal networks, allowing them to use compromised devices without their owners' knowledge.
For advertisers, one of the most concerning findings was how these networks used ordinary household internet connections to disguise malicious traffic.
This article examines how the XCPE botnet operated, why residential proxy traffic is difficult to identify, and what the incident means for businesses running paid advertising campaigns.
What is the XCPE botnet?
XCPE was a botnet consisting primarily of compromised Android TV set top boxes.
A botnet is a network of devices infected with malicious software that allows an operator to control them remotely.
In many cases, device owners remain unaware of the infection.
According to Google's investigation, Videostrong installed a hidden backdoor in devices during manufacturing.
This backdoor allowed operators to install additional malware and provide access to other criminal networks.
These networks used the compromised devices to:
- Generate fraudulent advertising activity.
- Route internet traffic through residential connections.
- Install additional malware without the owner's knowledge.
- Access and steal sensitive information.
- Execute commands remotely.
Google found the malware across different device models and brands over several years.
Although most infected devices were Android TV boxes, researchers also identified infections involving other Android devices, including telehealth equipment, karaoke machines, and digital signage.
The operation demonstrates how consumer devices might become part of an advertising fraud network without any action from their owners.
How did the XCPE botnet generate advertising fraud?
One of the malicious activities associated with XCPE involved advertising fraud.
The botnet allowed criminal affiliates to install software modules that performed different activities on compromised devices.
Some modules supported advertising fraud, while others turned devices into residential proxies.
These activities gave criminal networks access to devices and internet connections belonging to ordinary users.
1. Generating fraudulent advertising activity
Advertising fraud occurs when malicious actors generate invalid advertising interactions to manipulate advertising systems or obtain financial benefits.
The XCPE investigation identified malware modules designed to conduct advertising fraud.
These modules allowed criminal networks to generate fraudulent activity through compromised devices without their owners' involvement.
For advertisers, fraudulent clicks and other invalid interactions create concerns about advertising efficiency and the quality of campaign traffic.
2. Using residential internet connections
The botnet also allowed criminal networks to route traffic through infected devices connected to residential networks.
This is particularly concerning because residential traffic often appears more legitimate than traffic originating from known data centres.
A fraudulent request routed through a household connection might appear to originate from an ordinary internet user.
The IP address might belong to a legitimate internet service provider, and its geographical location might match an advertiser's campaign targeting.
These details make it harder to identify suspicious activity through basic IP checks alone.
Why residential proxy traffic is difficult to identify
Advertisers often examine traffic sources, IP addresses, geographical locations, and visitor behaviour when assessing campaign performance.
These details provide useful information, but they do not always establish whether an interaction comes from a genuine potential customer.
Residential proxies create an additional challenge.
A residential proxy routes internet traffic through an IP address associated with a household internet connection.
Because these addresses belong to residential networks, the resulting traffic might resemble ordinary website visits.
Consider an ecommerce business running Google Ads campaigns targeting customers in Germany.
A fraudulent request might pass through a compromised device connected to a household internet connection in Germany.
The traffic might display:
- A residential IP address.
- A location within the advertiser's target market.
- An internet service provider associated with ordinary households.
From these details alone, the interaction might appear legitimate.
But the person who owns the internet connection might have no knowledge of the activity.
The request might originate from software controlled by a criminal network rather than someone interested in purchasing the advertised product.
This is why a residential IP address does not automatically indicate genuine customer interest.
What does residential proxy ad fraud mean for advertisers?
The XCPE case highlights several challenges for ecommerce and lead generation businesses running their own advertising campaigns.
1. Wasted advertising budgets
When fraudulent activity results in billable ad clicks, advertisers risk paying for interactions with no genuine buying intent.
These clicks consume money intended to attract potential customers.
For ecommerce businesses, the concern is whether advertising traffic has a realistic chance of generating purchases.
For lead generation businesses, the concern is whether paid clicks come from genuine prospects who might become paying customers.
Fraudulent interactions contribute neither.
2. Difficulty identifying suspicious traffic
Residential proxy networks make traffic identification more complicated.
An advertiser might see clicks from expected locations and familiar internet service providers without immediately noticing anything unusual.
This creates a risk of assuming traffic is legitimate based on geographical location or IP address alone.
The XCPE operation demonstrates why advertisers need to consider visitor behaviour alongside basic traffic information.
3. Misleading campaign performance data
Fraudulent traffic might increase reported website visits and advertising interactions without generating meaningful business results.
For example, an ecommerce campaign might receive additional clicks without a corresponding increase in purchases.
A lead generation campaign might record website activity without attracting qualified prospects.
These patterns might make it harder for advertisers to understand why campaigns are underperforming.
Although poor conversion performance does not automatically indicate click fraud, suspicious traffic deserves further investigation.
How did Google dismantle the XCPE botnet?
Google worked with Japan's National Police Agency and several security organisations to investigate and disrupt the operation.
The company took legal action against the infrastructure used to control infected devices and distribute malicious software.
According to its November 2026 report, Google:
- Disrupted the domains and servers used to operate the XCPE botnet.
- Targeted infrastructure associated with additional malware distributed through the compromised devices.
- Ended its relationship with Videostrong Technology.
- Banned Videostrong and its affiliates from producing certified Android devices or supplying software to certified manufacturers.
- Introduced improvements to its device certification and security review processes.
Google also reported that the disruption of infrastructure associated with the malware affiliates was expected to cut off access to at least 10 million additional compromised devices.
These devices were separate from the more than 1.3 million devices enrolled in the XCPE botnet.
The takedown represents a significant disruption to the operation, although it does not eliminate the wider threat of residential proxy abuse and advertising fraud.
How ClickPatrol helps advertisers identify suspicious traffic
The XCPE incident demonstrates why advertisers should not rely solely on IP addresses or geographical locations to determine whether website traffic is legitimate.
Fraudulent activity might originate from residential connections and appear similar to ordinary website visits.
At ClickPatrol, our traffic protection algorithm evaluates visitor behaviour and suspicious activity to identify potentially fraudulent clicks.
Rather than relying on an IP address being associated with a particular location or network type, our system analyses traffic patterns to determine how interactions should be handled.
When ClickPatrol identifies fraudulent activity, it automatically blocks the traffic and applies the relevant exclusions to supported advertising campaigns.
This helps advertisers protect their budgets from unwanted clicks while preserving traffic from genuine potential customers.
Protect your advertising budget from fraudulent traffic
Google's dismantling of the XCPE botnet disrupted an operation involving more than 1.3 million infected devices.
But the case also highlights a wider challenge for advertisers.
Fraudulent advertising activity does not always originate from suspicious data centres or unfamiliar locations.
Sometimes, it comes through ordinary household internet connections.
For businesses investing in paid advertising, the important question is whether their clicks come from genuine potential customers.
A residential IP address does not guarantee legitimate traffic, and a click from an expected location does not automatically represent buying intent.
Start your free ClickPatrol trial to identify suspicious clicks and protect your advertising budget.
