Targeted ads delivered 0-click exploits to install mercenary malware

Researchers at the University of Toronto’s Citizen Lab have proved that malicious actors can embed zero‑click exploits in online advertisements, turning ordinary ad impressions into silent infection vectors for sophisticated spyware. The technique bypasses any user interaction, compromising iOS and Android devices the moment a page loads or an app opens. For anyone who relies on digital tools for work or personal communication, the threat reshapes the baseline of what “safe browsing” means.

Zero‑Click Exploits in the Advertising Supply Chain

The campaign, dubbed AdFection by Citizen Lab, leveraged a flaw that activates without a tap, swipe, or download, meaning the victim’s device is compromised merely by rendering an ad. This contrasts with traditional drive‑by attacks that require the user to click a malicious link or open a compromised file. By embedding the exploit in the ad payload, attackers exploit the trust placed in ad networks and the automatic rendering pipelines of mobile browsers and apps.

Technical analysis shows the exploit chains together a series of vulnerabilities in the operating system’s media handling components, allowing code execution at the kernel level. Once the malicious code runs, it installs a “mercenary” spyware suite capable of exfiltrating audio, video, keystrokes, and location data. The stealth is reinforced by the malware’s use of encrypted communication channels that blend with normal network traffic, evading standard intrusion‑detection signatures.

Because the infection vector is embedded in the ad delivery layer, conventional endpoint protections that focus on user‑initiated downloads miss the attack entirely. Only forensic hardware or specialized software that can capture low‑level system calls can reveal the presence of the payload. This shifts the defensive focus from the user’s device to the integrity of the advertising ecosystem itself.

The AdFection Campaign: Scope and Targets

Citizen Lab confirmed that 45 devices belonging to the international news outlet Al Jazeera were compromised during 2024 and 2025. The victims included journalists, editors, and other newsroom staff, indicating that the attackers were not merely interested in broad data harvesting but in precision targeting of media professionals. The campaign’s geographic spread suggests a coordinated effort, likely backed by a state or well‑funded criminal group seeking to monitor dissenting voices.

Both iOS and Android platforms were affected, demonstrating the attackers’ ability to develop cross‑platform exploits that exploit similar media‑processing pathways on each operating system. The dual‑platform reach expands the potential victim pool, as many newsrooms employ a mix of devices for flexibility and redundancy. The fact that the campaign persisted across two years without detection underscores the difficulty of spotting zero‑click infections in real time.

The use of “mercenary” malware—software sold or leased to clients for surveillance—implies a business model where the exploit code is licensed to multiple operators. This commoditization lowers the barrier for hostile actors to launch sophisticated espionage campaigns without developing their own zero‑click exploits. The ad‑based delivery mechanism further democratizes access, allowing attackers to reach targets without needing direct phishing or social engineering.

Implications for Media Organizations and Civil Society

Citizen Lab warns that the distribution of spyware via targeted ads poses a security risk not only to individual victims but also to entire states and civil‑society groups that rely on free press. When a newsroom’s devices are silently compromised, confidential sources, unpublished stories, and editorial strategies become exposed, eroding journalistic independence. The covert nature of the attack means that even organizations with strong cybersecurity policies may remain unaware of the breach.

Beyond the immediate loss of confidentiality, the campaign highlights a systemic vulnerability: the ad‑tech supply chain is an under‑examined attack surface. Advertisers, publishers, and ad‑network operators typically prioritize revenue and user experience over rigorous code‑signing or sandboxing of ad content. This creates a trust gap that malicious actors can exploit, turning a commercial ecosystem into a conduit for espionage.

For civil‑society organizations that depend on secure communications to coordinate activism or provide services, the threat expands the risk profile of any online activity. An undetected infection can feed real‑time intelligence to adversaries, enabling pre‑emptive repression or targeted disinformation. The broader implication is a shift from protecting data at rest to securing the entire data‑in‑motion pipeline, including the seemingly innocuous ad layers.

What This Actually Means For You

1. Zero‑click attacks bypass user vigilance. Even the most security‑conscious individuals can be infected without clicking anything.
2. Standard antivirus solutions are unlikely to detect the malware; only specialized forensic tools can uncover it.
3. Media professionals and activists should treat any ad‑served content as a potential attack vector, not just phishing links.
4. Organizations must audit their ad‑tech partners and consider sandboxing or blocking third‑party ads on critical devices.
5. Regular, independent security audits become essential to identify hidden compromises that conventional monitoring misses.

Immediate Action Steps

Begin by restricting the loading of third‑party advertisements on devices used for sensitive work. This can be achieved through network‑level ad blockers, secure browser configurations, or by employing mobile device management (MDM) policies that whitelist only approved apps and domains. Reducing exposure to ad content dramatically lowers the chance of encountering a zero‑click payload.

Simultaneously, schedule a forensic review of any device that accesses news‑related platforms or handles confidential material. Engage a security firm capable of low‑level memory analysis and kernel‑level inspection, as these are the only methods that can reliably detect the stealthy spyware described in the AdFection findings.

Frequently Asked Questions

How can a zero‑click exploit install malware without any user interaction?

The exploit takes advantage of a vulnerability in the operating system’s media processing code that runs automatically when an ad is rendered, allowing code execution without a click or download.

Why are traditional antivirus tools ineffective against the AdFection campaign?

The malware operates at the kernel level and uses encrypted communications that blend with normal traffic, so it evades signatures that standard antivirus products rely on.

What steps can news organizations take to protect their staff from ad‑based spyware?

Organizations should limit third‑party ad exposure on work devices, implement strict MDM policies, and conduct regular forensic audits to detect hidden infections.

What Do You Think?

Given that a simple ad impression can now serve as a silent entry point for state‑grade spyware, should newsrooms overhaul their entire digital workflow to treat every external script as a potential threat?