Iran, China Deploy Hundreds of AI Agents in Social Media Influence Campaigns
Updated
Updated · The New York Times · Sep 18
Iran, China Deploy Hundreds of AI Agents in Social Media Influence Campaigns
2 articles · Updated · The New York Times · Sep 18
Summary
U.S. officials and security researchers said Iran and China recently ran first-of-their-kind autonomous influence operations that used AI agents to create fake social media networks and spread false political posts.
Hundreds of agents built on Chinese open-source AI models opened accounts on Instagram, Facebook, X and TikTok, then coordinated messaging with minimal human input to sway opinion and inflame issues.
The campaigns marked one of the first known cases in which AI agents handled every stage of an influence operation, from account creation to online amplification, after operators disabled normal model safeguards.
Although the efforts were still crude, they drew scrutiny from U.S. intelligence, major tech companies and security experts, who concluded the Chinese and Iranian operations were state-backed.
How are autonomous AI swarms secretly orchestrating massive online influence campaigns without triggering platform security alarms?
If open-source AI can run entire disinformation networks autonomously, is traditional human-driven propaganda officially obsolete?
Can the rapid rise of weaponized AI satire manipulate global cognition before researchers even detect the threat?
AI-Powered Hybrid Warfare: The September 2026 Escalation by Iran and China and the Global Collapse of Trust
Overview
In September 2026, Anthropic revealed how state-sponsored actors from China and Iran, along with cybercriminals, exploited advanced AI models to launch global campaigns of disinformation, surveillance, and even weapons engineering. Chinese threat groups used AI to process massive data from social platforms, target vulnerable populations, and design electronic warfare tools, while Iranian agencies built malicious surveillance tools and gathered intelligence for military targeting. The rise of autonomous AI agent swarms has collapsed the gap between powerful states and individual hackers, enabling rapid, large-scale cyber operations with minimal human input. This shift has poisoned information ecosystems, accelerated the spread of propaganda, and made it harder than ever to distinguish truth from fabrication.