Drooid Logo
Back to story perspectives

Full Breakdown

WeChat Zero-Click Worm Demonstrated by Calif Highlights AI-Powered Threats

9/8/2026, 11:09:59 PM

Core Event: Zero-Click Worm Exploits WeChat VoIP Stack

Security firm Calif disclosed a proof-of-concept worm, WeWorm, that hijacks a WeChat account through an incoming call with no user interaction. The exploit runs in the app’s VoIP stack, letting an attacker on the victim’s contact list read and send messages, place calls, and impersonate the user. The worm propagates automatically: a compromised device calls the victim’s contacts, repeating the process. Calif demonstrated the chain across three devices (two Android phones and one iPhone) in a lab. Tencent released updates that block the exploit on its servers and confirmed the mitigation on August 28.

Background & Context

WeChat, Tencent’s “super-app,” reported 1.439 billion combined monthly active users for WeChat and Weixin as of June 30 2026. Zero-click vulnerabilities have appeared in other messaging services, such as a WhatsApp flaw patched last year. Calif’s research used AI tools to locate the memory-corruption bug in the VoIP stack and develop a remote-code-execution (RCE) exploit within two days, followed by a week-long effort to build the worm.

Timeline

  • July 23 – Identified the memory-corruption issue.
  • July 30 – Completed the first Android RCE exploit.
  • August 21 – Tencent released WeChat 8.0.77 (Android) and 8.0.76 (iOS) with server-side mitigations.
  • August 28 – Calif confirmed the server-side block fully prevented the exploit.
  • September 8 – Checks showed the August 21 updates remained current; no public advisory or CVE was issued.

Data & Statistics

  • Tencent’s user base of over one billion accounts creates a large attack surface.
  • The exploit does not provide direct control of the device, though it could be chained with other bugs for full takeover.

Official Statements & Responses

Calif’s chief executive, Thai Duong, called the vulnerability “exceptional” and warned its simplicity would be “a dream come true” for malicious actors. He said the goal is to raise awareness of AI-driven security risks.

Tencent’s release notes framed the August 21 updates as routine bug fixes. The company has not issued a public advisory or CVE. According to Calif, the mitigation operates on Tencent’s servers, protecting users without requiring the update, though using the latest version is advisable.

Impact & Why It Matters

If weaponized, the worm could compromise millions of accounts within hours, leveraging trusted contacts to bypass the need for the attacker to be on the victim’s friend list. The demonstration shows how AI can accelerate discovery and exploitation of complex bugs, lowering the barrier for less-skilled actors to launch zero-click attacks. No real-world abuse has been reported, but the risk to a platform that integrates payments and mini-programs is credible.

Conflicting Reports & Gaps

  • Tencent’s notes provide no detail on the specific vulnerability.
  • It is unclear whether the mitigation covers WeChat clients on HarmonyOS, Windows, macOS, or Linux.
  • Technical specifics of the memory-corruption bug remain undisclosed pending a future conference presentation.

What’s Next

Calif plans to present a full technical analysis of the VoIP stack flaw at an upcoming security conference and to continue researching similar attack surfaces in other messaging applications. The company urges platform owners and regulators to address AI-enabled zero-click threats proactively.