# Researchers used AI to build a WeChat worm that spreads through phone calls

> **Open Intelligence Dossier** · First detected: 2026-09-11 16:07 UTC · Category: Technology

## Executive Summary
Researchers have demonstrated a new AI-powered worm capable of infiltrating WeChat and spreading through phone calls, signaling a shift in cyber threats.

## Intelligence Brief
Security researchers have successfully developed a specialized worm targeting the WeChat platform. According to reporting from The Next Web, this particular malware utilizes artificial intelligence to facilitate its distribution. A critical component of this discovery is the mechanism of transmission; the coverage specifies that the worm is designed to spread through phone calls. This represents a departure from traditional digital infection vectors, integrating AI capabilities to move across a communication network that includes voice interactions. Coverage of this event is highlighted by The New York Times and The Next Web. The New York Times emphasizes the broader implications of the attack, describing the simulation as a mock AI attack.


The publication suggests that this demonstration signals the beginning of a dangerous new era for digital security in China. Both outlets focus on the technical feasibility of using AI to automate the spread of malicious code within one of the world's largest social messaging ecosystems, showcasing the potential for rapid, automated proliferation. To understand the significance of this development, it is necessary to note the central role of WeChat in daily communication and commerce. By utilizing AI to weaponize phone calls, the researchers have shown that traditional barriers to worm propagation can be bypassed. This specific context is vital because it demonstrates that AI is not merely being used for data analysis or content generation, but as a tool for constructing active, self-propagating cyber weapons. The transition from text-based or link-based phishing to voice-integrated AI worms marks a technical evolution in how malware operates.


Future developments to monitor include how WeChat or its parent company responds to these findings. While the coverage establishes that this was a mock attack conducted by researchers, it does not specify if a patch has been deployed or if similar vulnerabilities exist in other messaging apps. Observers will be looking for official statements regarding the security of voice-call integrations. The primary focus remains on whether the techniques used in this AI-driven worm can be replicated by malicious actors for actual wide-scale deployment across the platform.

## Multi-Source Evidence Table
| Source Outlet | Headline | Verification URL |
|---|---|---|
| The New York Times | For China, a Mock A.I. Attack on WeChat Signals a Dangerous New Era | [Source Link](https://news.google.com/rss/articles/CBMifkFVX3lxTFBLT1V6dG0xRGNxMDBjbF9KN3VvUlcwU2Uxa21PYVY5Wm1ET3dqODJOcmlEQUlCbjloZENlY1pSUnJmdU5BWGZ4cWdUeEUzSXZGU0trbEdwWGM0eHh5NkQ3RE04UDN2am1tVzFPaDdRLTJlLW81dWUwLUt1Z0Jidw?oc=5) |
| The Next Web | Researchers used AI to build a WeChat worm that spreads through phone calls | [Source Link](https://news.google.com/rss/articles/CBMidkFVX3lxTE9FZmt6RFltLWNibDNBQUt5ZmZmNWdBaVUtMG1GVXNoQ1RCXzRQeFlVczVEeTN4MXc4T2VTT1RaeVRPZk53bDM2QUVBMnBFNlkwRVN2T3NWOHEtTXc0QnlHMkpZa2JURlVsSGZmemh1TDhpTWFNaFE?oc=5) |

---
*Canonical Source: https://pulse.byoviral.com/trend/2026-09-11/researchers-used-ai-to-build-a-wechat-worm-that-spreads-through-phone-calls*