Home / Blog / BADBOX 3.0 botnet returns with record-breaking 40 billion…
Tech News

BADBOX 3.0 botnet returns with record-breaking 40 billion daily ad requests

The BADBOX 3.0 botnet has resurfaced across infected Android streaming devices to generate a record forty billion fraudulent advertising bid requests per day.

By Dillip Chowdary • Nov 21, 2026 • Source: BleepingComputer

BADBOX 3.0 botnet returns with record-breaking 40 billion daily ad requests
Image: BleepingComputer

At a glance

Threat name
BADBOX 3.0
Daily bid requests
40 billion fraudulent requests
Connected TV share
Over 4 percent of traffic
Infection scale
Over 10 million Android devices
Associated proxy
PXYEDGE (formerly IpMoYu)
Peak proxy connections
Approximately 340,000 nodes

Security researchers at HUMAN's Satori threat intelligence team uncovered and disrupted BADBOX 3.0, a revived malware botnet generating 40 billion fraudulent advertising bid requests daily across infected Android devices. As detailed in BleepingComputer's report, the operation has grown into the largest documented digital ad-fraud scheme in history following previous takedowns of earlier infrastructure variants.

This report details the technical mechanics behind the botnet, its hardware infection chain, the multi-network residential proxy architecture, and the multi-agency disruption campaigns targeting control servers and repositories. It is written for cybersecurity teams, streaming media operators, ad fraud analysts, and consumers managing off-brand connected television hardware.

01Record bid volume and connected television impact

At its operating peak, the BADBOX 3.0 cluster generated roughly 40 billion fraudulent advertising bid requests daily. The Human Defense Platform observed that this volume represented over 4% of all connected television traffic monitored across its global sensors. Researchers confirmed the operation as the largest documented ad-fraud botnet by daily bid volume.

Traffic generation from this variant expanded to more than three times the scale of VASTFLUX, the previous record holder for automated ad fraud. The daily peak activity exceeded the combined peak volumes documented across the last five years of published Satori investigations, severely stressing ad exchange filters.

Digital ad networks encountered spoofed traffic volumes eclipsing all prior benchmarks observed by security researchers. The botnet leveraged compromised endpoints to overwhelm verification pipelines across ad exchanges without raising immediate alarms. Operators converted commodity consumer hardware into an automated monetization engine running continuously in the background.

| Metric | Predecessor or Stated Baseline | BADBOX 3.0 Observed Level | | Peak daily ad requests | VASTFLUX record | 40 billion bid requests | | Connected TV share | Baseline network activity | Over 4% of platform traffic | | Investigation comparison | Five years of Satori studies | Surpasses combined peaks | | Proxy network nodes | 200,000 advertised on PXYEDGE | 340,000 peak unique connections |

02Retail hardware origins and enforcement history

The BADBOX malware was originally discovered in early 2023 on an off-brand T95 Android TV box purchased on Amazon. By December 2024, German cybersecurity agencies warned that the botnet had infected tens of thousands of consumer streaming devices. The hardware was sold directly to consumers without standard integrity protections.

Disruption efforts escalated in March 2025 when cybersecurity firm HUMAN and the Shadowserver Foundation disconnected more than 500,000 BADBOX 2.0 endpoints from control servers. Simultaneously, Google removed 24 malicious loader applications from the Google Play store that had served as distribution vectors for the rogue code.

Google subsequently sued the operators, disclosing in court filings that BADBOX 2.0 had infected over 10 million Android devices by April 2025. Despite facing federal litigation and infrastructure takedowns, the threat group redesigned its architecture to deploy the more resilient BADBOX 3.0 iteration.

03Hidden video playback and streaming impersonation

Advertisement

Tech Pulse Daily

Get tomorrow's pulse first

Join engineers who read Tech Pulse before stand-up. Free, weekday mornings.

The operational method relies on compromising low-cost streaming media boxes, mobile phones, and tablets powered by the Android Open Source Project. Threat actors target generic AOSP hardware because these systems frequently lack standard enterprise device certifications and proprietary firmware safeguards. Once active inside the operating system, the payload executes ad-fraud scripts alongside background traffic proxying routines.

To generate fraudulent advertising revenue, the malware launches automated video advertisements inside hidden interface windows that remain entirely invisible to the device owner. The software concurrently impersonates legitimate streaming applications to make outgoing bid requests appear authentic to exchange verification algorithms. Viewers watching regular programming remain unaware that high-volume commercial playback runs continuously beneath active application layers.

Investigators also tracked sophisticated fraudulent server-side ad insertion schemes orchestrated by the botnet controllers. These SSAI requests deliberately tricked ad buyers into believing promotional videos were streaming onto connected television hardware when no real devices or video displays were involved. Advertisers paid premium rates for phantom impressions delivered entirely through automated server manipulations.

04Residential proxy enrollment across eleven exit networks

In addition to ad fraud, a specialized proxying module silently enlists infected consumer hardware into one of 11 residential proxy networks. Device owners never receive notification or provide consent when their home internet connections are converted into distributed commercial network relays. The operators monetize both fraudulent ad clicks and third-party bandwidth routing through the same physical devices.

Security analysts identified one prominent routing cluster, named PXYEDGE, as the rebranded successor to IpMoYu, a residential proxy operation previously tied to BADBOX 2.0 infrastructure. While PXYEDGE promotional listings advertised approximately 200,000 residential exit nodes to customers, live network telemetry monitored by researchers revealed unique concurrent connections peaking at roughly 340,000 endpoints.

The botnet operators route a significant portion of their own fraudulent advertising traffic directly through these hijacked residential proxies. Bouncing fake bid requests through domestic residential IP ranges conceals malicious automation from fraud detection systems and anti-abuse filters. This layered proxying approach makes artificial connected TV impressions appear indistinguishable from legitimate household viewing patterns.

05Factory firmware roots and coordinated domain seizures

Forensic investigations traced the malware back to a specific Chinese electronics manufacturer by inspecting domain registrations, device-specific code routines, factory-installed TV launchers, and associated repository maintainers. The infected hardware arrives from the factory preloaded with components designed to pull secondary malicious modules from public code repositories. Payloads are also mirrored across redundant hosting providers to resist takedowns.

Disruption efforts resumed earlier this month when HUMAN collaborated with Google to dismantle the related XCPE malware network, seizing multiple command-and-control domains utilized by BADBOX 3.0. Law enforcement and security partners also dismantled an active malware code repository, while HUMAN deployed real-time network filtering to block fraudulent ad bids linked to the botnet.

Security researchers emphasize that eliminating firmware-level risks requires disciplined procurement practices and continuous baseline security verification across all household streaming hardware:

  1. Hardware procurement Consumers should only buy certified Android devices from reputable hardware manufacturers.
  2. Security configuration Device owners must verify that Google Play Protect remains actively installed and enabled.

Note: HUMAN warns that the botnet operators are likely to adapt to the latest disruption, as they have done in past enforcement rounds.

Frequently asked questions

What is BADBOX 3.0?

BADBOX 3.0 is an ad-fraud botnet infecting generic Android streaming devices to generate 40 billion fraudulent ad bid requests daily.

How does the malware commit ad fraud?

It plays hidden video ads while impersonating streaming apps and generating fake server-side ad insertion requests that trick advertisers.

What is the role of residential proxies?

Infected endpoints are enrolled across 11 residential proxy networks, including PXYEDGE, routing ad fraud through home connections to bypass filters.

How can consumers protect their devices?

Users should purchase certified Android devices from reputable manufacturers and verify that Google Play Protect remains enabled.

Sources

Dillip Chowdary

Author

Dillip Chowdary

Writes Tech Bytes coverage of AI, engineering, and the tools that actually ship. Editor of Tech Pulse Daily.

Related on Tech Bytes

Advertisement

5-min tech signal

Weekday briefing for engineers who skip the noise.

No spam · Unsubscribe anytime

Advertisement

✈️ CareerPilot

Your AI job-search copilot

Match your resume against live Ashby, Greenhouse & Lever openings — fit scores, job-specific resume optimization and email alerts.

Find matching jobs →

Free Tools

Browse all tools →