Zero-Click Spyware Delivered Through Online Ads Confirmed in First Known Cases

Citizen Lab researchers at the University of Toronto have documented what they describe as the first clear instance of mercenary spyware reaching targets through ordinary online advertising networks. The report details how an Israeli firm called Discover-Tech has operated as a conduit between ad vendors and companies selling zero-click surveillance tools. Those tools, sold to governments, can infect devices without any action from the user.

The lab identified two spyware vendors, Candiru and Bindecy, that routed campaigns through Discover-Tech’s platform. One confirmed operation delivered Candiru’s malware to at least 45 Windows machines belonging to Al Jazeera journalists and staff between 2024 and 2025. The infections occurred as employees browsed or used apps that served the malicious ads. Researchers named the technique AdFection to capture how commercial advertising infrastructure now doubles as a delivery system for state-linked hacking.

Zero-click attacks have long been prized because they leave almost no trace for victims to notice. Until now, the most sophisticated versions relied on flaws in messaging apps or other direct communication channels. Shifting the vector to ad networks changes the scale and the level of operational security required. An operator can specify an audience by location, device type, or browsing behavior and let the ad exchange handle the rest. The target never needs to click anything or receive a suspicious message.

Discover-Tech presents itself as a standard advertising technology company. Citizen Lab’s findings show it connected spyware operators to the global ad ecosystem in ways that allowed precise targeting while maintaining distance from the actual infection code. The lab stopped short of naming specific governments as clients, but the pattern matches known customers of Candiru and similar firms, which have included intelligence agencies in Europe, the Middle East, and elsewhere.

The implications extend beyond the immediate victims. Independent media outlets already operate under resource constraints and heightened scrutiny in many countries. When the same ad networks that finance legitimate journalism can be turned into infection points, the cost of doing sensitive reporting rises further. Staff at Al Jazeera were not chosen at random; they work for an organization whose coverage has repeatedly drawn the attention of state actors.

For ordinary users the risk is harder to quantify but no less real. The ad ecosystem reaches billions of devices daily. Even if operators intend to limit campaigns to high-value targets, the technical pathway is now public enough that others may copy it. Bill Marczak, a senior researcher at Citizen Lab, noted that someone compromised this way would see nothing unusual and might not even register that an ad had loaded.

The episode underscores how commercial surveillance infrastructure and state intelligence capabilities have become intertwined. Advertising platforms were built to collect detailed behavioral data for commercial purposes; that same data can now be repurposed to select espionage targets. Regulators have spent years examining the privacy harms of ad tech without confronting its potential role in enabling offensive cyber operations. The Citizen Lab report supplies concrete evidence that the two problems are no longer separable.