Researchers say they used AI to build a zero-click worm that can hack WeChat accounts and spread across iOS and Android; Tencent says it fixed the vulnerability
The attack, discovered by A.I. researchers, could have compromised hundreds of millions of devices within hours, experts said.
Context & Ripple Effects
The reported WeChat demonstration follows a June 2026 research claim that open-source AI could build a worm that adapts attacks to individual computers, moving AI-assisted malware from a single-host proof of concept toward self-propagation. It also extends the zero-click risk already exposed by WhatsApp’s patch for attacks against targeted iOS and Mac users to a cross-platform messaging service.
Tencent says it has fixed the reported flaw, but the research matters because a call-based, no-interaction compromise combines account takeover with automated contact-to-contact distribution across iOS and Android.
First-order effects
- Tencent’s fix removes the reported WeChat entry point, while the researchers’ demonstration gives defenders a concrete cross-platform zero-click attack chain to test against.
- WeChat users are no longer dependent on answering a call for the reported attack to begin; Tencent’s remediation becomes the immediate protection boundary.
Second-order effects
- Tencent and other messaging-platform operators face pressure to audit incoming-call and account-recovery paths for flaws that can be chained into autonomous propagation, rather than treating zero-click bugs as isolated device compromises.
- The finding reinforces the operational burden on iOS and Android security teams: a messaging-service weakness can create exposure on both platforms even when the exploit originates outside the operating system.
Third-order effects
- If AI-assisted development keeps reducing the effort needed to assemble exploit chains, defenders will need to prioritize vulnerabilities by propagation potential, not only by the severity of a single account compromise.
- The pattern points toward an agentic attack surface in which broadly distributed communications services become the highest-leverage targets because compromise can supply the next set of victims automatically.
The trend: AI-assisted offensive tooling is converging with zero-click messaging flaws, raising the risk that account compromises become self-propagating cross-platform incidents.