The Brief
Society & Ethics 4 min read

Ads as Attack Vectors: The Zero-Click Spyware Threat

NAVION

Share

Most people understand, at least vaguely, that online ads track them. Fewer understand that an ad can now silently install spyware on a device without the user clicking anything, seeing anything unusual, or making any mistake. Researchers at The Citizen Lab at the University of Toronto have documented exactly this scenario, naming it “AdFection” and describing it as the first detailed technical evidence made publicly available of sophisticated spyware being delivered through digital advertising infrastructure.

How an Ad Becomes a Weapon

To understand AdFection, it helps to understand how online advertising actually works. When a user opens an app or visits a website, an automated auction runs in milliseconds to determine which ad gets displayed. This process, known as real-time bidding, involves advertisers collecting substantial amounts of data about the device, its location, and browsing history. The system is built for precision targeting. That same precision, it turns out, can be repurposed for something far more dangerous.

Citizen Lab’s researchers found that an Israeli-registered advertising firm called Discover-Tech allegedly connected its servers to infrastructure linked to Candiru, an Israeli spyware developer. Discover-Tech operates an advertising platform called Controlium, which the researchers say can allow its users to de-anonymize, fingerprint, and automatically infect devices once an ad appears on screen. The infected device does not need to interact with the ad. The user does not need to tap, click, or even notice anything. The attack is what security researchers call “zero-click”: fully automated, fully invisible.

Citizen Lab estimates that millions of devices were directed to Candiru’s servers through this mechanism. By analyzing ad-bidding records that were publicly exposed on Discover-Tech’s own website, the researchers found that ads linked to the firm appeared in more than 70,000 mobile apps and more than 120,000 websites between March and June of this year. Multiple apps displayed more than a million ads connected to Discover-Tech.

45 Devices, One News Outlet, and a Global Footprint

The clearest documented case involves Al Jazeera. During 2024 and 2025, Citizen Lab concluded that 45 devices belonging to the news outlet were targeted with zero-click attacks delivered through ads. Al Jazeera participated in the investigation and shared device logs with the researchers. Those logs were, according to Citizen Lab senior researcher Bill Marczak, essential to understanding the ad-based nature of the attack. Without that level of network monitoring, the attack would likely have gone undetected entirely.

That visibility problem is significant. Marczak notes that if a user’s personal device were targeted while connected to home Wi-Fi or mobile data, there would be no equivalent monitoring in place. The absence of prior public reporting on this type of attack is, in his assessment, mainly a visibility issue: defenders need to be in the right place, at the right time, looking at the right data.

The researchers also observed Discover-Tech ad impressions on government and national-security-linked networks around the world, including NATO and United Nations networks, lawmaking bodies in Europe, critical infrastructure providers in Australia, and armed forces and government agencies in Norway and Sweden. Not all of those impressions were necessarily weaponized, but the reach illustrates how deeply advertising infrastructure is embedded in sensitive digital environments.

Following Citizen Lab’s findings, Google terminated Discover-Tech’s registration from its ad platforms and set up Safe Browsing warnings for websites linked to the group. The Interactive Advertising Bureau’s European branch suspended Discover-Tech from its global vendor list. Beeswax, an advertising industry company owned by Comcast, also removed the firm from its platform.

Why This Goes Beyond a Single Campaign

The AdFection case is not primarily a story about one bad actor. It is a story about structural vulnerability. The online advertising ecosystem was built for reach, speed, and targeting precision. Those properties make it extraordinarily useful for legitimate advertisers. They also make it, as Marczak puts it, “perfectly suited to the mercenary spyware industry.”

Zero-click attacks have existed before, typically exploiting vulnerabilities in messaging apps. What AdFection represents is a convergence: the stealth of zero-click malware combined with the ubiquity and trusted status of digital advertising. Ads are everywhere. They are embedded in apps people use for news, navigation, entertainment, and communication. Blocking them thoroughly across all platforms is, according to Citizen Lab, one of the few practical defenses available to ordinary users, and even that is imperfect.

The deeper issue is one of accountability. Advertising infrastructure operates at enormous scale, with many intermediaries between an advertiser and the device that ultimately displays an ad. That complexity creates gaps where malicious actors can operate. Until ad companies implement serious vetting of the entities they work with, or until the underlying architecture changes, the conditions that enabled AdFection remain in place.

In Short

An ad appearing on a screen can now deliver spyware to a device without any action from the user. Citizen Lab documented this happening to 45 devices at Al Jazeera, with ads traced to a firm called Discover-Tech and spyware attributed to Candiru. The attack exploits the same targeting infrastructure that makes digital advertising commercially valuable. The scale of that infrastructure, and the trust placed in it by apps, websites, and even sensitive government networks, is precisely what makes this threat difficult to contain.

Based on reporting from Wired.

Written by

NAVION