Security researchers have discovered a large-scale distribution network for spyware that compromises iPhones. At least 6,600 legitimate websites have been made to load malicious code without their operators' knowledge, and 248 of them serve as "direct delivery" hubs with a combined monthly audience of roughly 8 million visitors. Users do not even need to click a suspicious link — the attack begins the moment an ad appears on screen.
The findings come from an investigation published on October 22, 2026 (US Eastern Time) by mobile security firm iVerify and cybersecurity firm Report URI. The two companies identified two attack campaigns run by a criminal group believed to have ties to China.
The attack begins the instant an ad is displayed
Rocky Cole, co-founder and chief operating officer of iVerify, described the nature of the technique: "The user isn't visiting a malicious site. They're not doing anything wrong. They're just scrolling, and they get infected."
iVerify's investigation confirmed a case in which an iPhone belonging to an employee at a financial institution was infected after the user scrolled past a soccer ad displayed on Instagram. The malicious code activated the moment the ad appeared on screen, identified the device's OS, and deployed the matching attack code. It stole login credentials and searched for cryptocurrency wallets.
The other campaign was caught by Report URI. The attackers acquired 630 expired domains that legitimate sites continued to rely on as sources for loading chatbots and analytics tools. The number of sites whose domains were hijacked reached roughly 8,000, which redirected traffic from vulnerable iPhones to spyware. The criminals' objective was to steal funds from cryptocurrency wallets and capture authentication credentials.
Scott Helme, founder and chief executive officer of Report URI, disclosed the following breakdown:
Note: Figures are tallied as of the iVerify and Report URI investigation. Two separate counts coexist because of differences in investigation methodology.
"DeadTag" suspected of selling delivery slots
Helme believes a group he has dubbed "DeadTag" is selling opportunities to deliver malicious content — whatever other criminals want — to people who visit legitimate sites without realizing they have been compromised. The group reportedly prepares up to 10 new domains loaded with malicious code each week. DeadTag's customers include criminal groups that specialize in compromising iPhones.
The criminal groups are using two types of "attack toolkits" that were reportedly used by state intelligence agencies to breach iPhones and later leaked. Cybercriminals have been operating both toolkits since spring 2026. Apple has already patched the vulnerabilities exploited in past attacks, but the attackers are deliberately targeting devices that have not updated their OS, where the pre-patch flaws remain.
Helme also said he has found evidence that the attackers are preparing to switch to code that exploits unpatched iOS vulnerabilities. Report URI is attempting to notify compromised sites, but the sheer number of targets has made the work difficult, and the firm is seeking cooperation from national cybersecurity agencies.
The findings show that spyware is shifting from a tool aimed at a select few high-value targets to a weapon of everyday cybercrime. iVerify reported last year on evidence of the first spyware campaign targeting iPhones in the United States and the European Union. The University of Toronto's Citizen Lab disclosed on October 15, 2026, that a Windows PC used by an Al Jazeera journalist was attacked via targeted advertising. At least 45 devices were reportedly affected, and an advertising technology company is suspected of cooperating with the espionage operation.
The reality of updates and ad-based intrusion
Neither Cole nor Helme has established the attackers' origin with certainty. Helme would say only that "they are Chinese, or at least people who want us to think so." The program code used in the attack, however, was written in Chinese. Forbes has also confirmed evidence that several domains used by the attackers blocked external cyberattacks using a China-based web security service.
As of publication on October 22, US Eastern Time, neither Apple nor Meta, which operates Instagram, had responded to requests for .
The market impact lies in the erosion of the very security assumptions underpinning mobile devices. Until now, phishing countermeasures have centered on "don't open suspicious links," but attacks routed through ad delivery networks and expired domains involve no user action at all. Because ad platforms can narrow their targeting by interests and device characteristics, attackers gain the advantage of being able to single out and target only vulnerable devices.
For corporate IT departments, the risk of BYOD environments — where the OS update status of personally owned devices cannot be fully tracked — is surfacing once again. Businesses that handle cryptocurrency wallets as part of their operations and media companies that outsource ad inventory may be forced to audit their delivery channels and take stock of the domains they depend on.
Both Cole and Helme point out that the attacks depend on vulnerabilities in specific software versions, and they stress the importance of keeping mobile operating systems up to date.
The full story
This article is one source in a clustered incident — the cluster page carries the summary, timeline and every other outlet covering it.
