Cybersecurity researchers have flagged multiple in-the-wild exploit campaigns that leveraged now-patched flaws in Apple Safari and Google Chrome browsers to infect mobile users with information-stealing malware.
“These campaigns delivered n-day exploits for which patches were available, but would still be effective against unpatched devices,” Google Threat Analysis Group (TAG) researcher Clement Lecigne said in a report shared with The Hacker News.
The activity, observed between November 2023 and July 2024, is notable for delivering the exploits by means of a watering hole attack on Mongolian government websites, cabinet.gov[.]mn and mfa.gov[.]mn.
The intrusion set has been attributed with moderate confidence to a Russian state-backed threat actor codenamed APT29 (aka Midnight Blizzard), with parallels observed between the exploits used in the campaigns and those previously linked to commercial surveillance vendors (CSVs) Intellexa and NSO Group, indicating exploit reuse.
The vulnerabilities at the center of the campaigns are listed below –
- CVE-2023-41993 – A WebKit flaw that could result in arbitrary code execution when processing specially crafted web content (Fixed by Apple in iOS 16.7 and Safari 16.6.1 in September 2023)
- CVE-2024-4671 – A use-after-free flaw in Chrome’s Visuals component that could result in arbitrary code execution (Fixed by Google in Chrome version 124.0.6367.201/.202 for Windows and macOS, and version 124.0.6367.201 for Linux in May 2024)
- CVE-2024-5274 – A type confusion flaw in the V8 JavaScript and WebAssembly engine that could result in arbitrary code execution (Fixed by Google in Chrome version 125.0.6422.112/.113 for Windows and macOS, and version 125.0.6422.112 for Linux in May 2024)
The November 2023 and February 2024 campaigns are said to have involved the compromises of the two Mongolian government websites – both in the first and only mfa.gov[.]mn in the latter – to deliver an exploit for CVE-2023-41993 by means of a malicious iframe component pointing to an actor-controlled domain.
“When visited with an iPhone or iPad device, the watering hole sites used an iframe to serve a reconnaissance payload, which performed validation checks before ultimately downloading and deploying another payload with the WebKit exploit to exfiltrate browser cookies from the device,” Google said.
The payload is a cookie stealer framework that Google TAG previously detailed in connection with the 2021 exploitation of an iOS zero-day (CVE-2021-1879) to harvest authentication cookies from several popular websites, including Google, Microsoft, LinkedIn, Facebook, Yahoo, GitHub, and Apple iCloud, and send them via WebSocket to an attacker-controlled IP address.
“The victim would need to have a session open on these websites from Safari for cookies to be successfully exfiltrated,” Google noted at the time, adding “attackers used LinkedIn messaging to target government officials from western European countries by sending them malicious links.”
The fact that the cookie stealer module also singles out the website “webmail.mfa.gov[.]mn” suggests that Mongolian government employees were a likely target of the iOS campaign.
The mfa.gov[.]mn website was infected a third time in July 2024 to inject JavScript code that redirected Android users using Chrome to a malicious link that served an exploit chain combining the flaws CVE-2024-5274 and CVE-2024-4671 to deploy a browser information stealing payload.
In particular, the attack sequence uses CVE-2024-5274 to compromise the renderer and CVE-2024-4671 to achieve a sandbox escape vulnerability, ultimately making it possible to break out of Chrome site isolation protections and deliver a stealer malware.
“This campaign delivers a simple binary deleting all Chrome Crash reports and exfiltrating the following Chrome databases back to the track-adv[.]com server – similar to the basic final payload seen in the earlier iOS campaigns,” Google TAG noted.
The tech giant further said the exploits used in the November 2023 watering hole attack and by Intellexa in September 2023 share the same trigger code, a pattern also observed in the triggers for CVE-2024-5274 used in the July 2024 watering hole attack and by NSO Group in May 2024.
What’s more, the exploit for CVE-2024-4671 is said to share similarities with a previous Chrome sandbox escape that Intellexa was discovered as using in the wild in connection with another Chrome flaw CVE-2021-37973, which was addressed by Google in September 2021.
While it’s currently not clear how the attackers managed to acquire the exploits for the three flaws, the findings make it amply clear that nation-state actors are using n-day exploits that were originally used as zero-days by CSVs.
It, however, raises the possibility that the exploits may have been procured from a vulnerability broker who previously sold them to the spyware vendors as zero-days, a steady supply of which keeps the ball rolling as Apple and Google shore up defenses.
“Moreover, watering hole attacks remain a threat where sophisticated exploits can be utilized to target those that visit sites regularly, including on mobile devices,” the researchers said. “Watering holes can still be an effective avenue for n-day exploits by mass targeting a population that might still run unpatched browsers.”
