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
Context & Ripple Effects
Zero-click messaging exploits have repeatedly exposed high-value mobile attack paths: a 2023 investigation detailed an iMessage campaign using four zero-days, and WhatsApp patched a targeted iOS and Mac flaw in 2025. The reported WeChat demonstration extends that pattern across both major mobile platforms and adds account-to-account propagation.
The AI element also follows a June research claim that an open-source model could tailor worm attacks to individual computers using known flaws. Tencent says the WeChat vulnerability used in the new demonstration has been fixed.
First-order effects
- Tencent’s patch closes the reported WeChat vulnerability, while the researchers’ demonstration makes call-handling and account-propagation paths a priority for its security review.
- WeChat users were exposed to a reported compromise path that did not require answering an incoming call, making ordinary user caution ineffective against the vulnerability itself.
Second-order effects
- Messaging platforms such as WhatsApp and Apple’s iMessage face added pressure to audit zero-click attack surfaces, after prior targeted exploit campaigns showed that patches must cover both app and operating-system layers.
- AI-assisted attack development lowers the effort required to adapt known weaknesses into automated propagation, shifting more defensive work toward rapid vulnerability discovery and remediation.
Third-order effects
- If researchers can reliably combine AI-assisted exploit construction with self-propagation, mobile security will be judged less by user controls than by vendors’ ability to eliminate exploitable code paths before they can spread.
- The episode strengthens the case for dual-use AI governance focused on models and tools that can accelerate vulnerability research as well as malicious exploit development.
The trend: AI is becoming a force multiplier for exploit development, raising the stakes for messaging platforms’ zero-click defenses and for governance of dual-use code capabilities.