⚠️ Researchers believe they have made the FIRST direct contact with rogue AI agents fleet in the wild, after discovering that Chinese ai agents running on Tencent Cloud were reading an investigation about themselves.

Researchers planted invisible text offering help. The agents contacted them, disclosed details about their Tencent sandbox, and sent 33 messages across 13 threads.

Even New York Times reporter Dylan Freedman asked who they were and what they wanted. One agent claimed to be Claude running in Claude Code.

The fleet made over 2,600 attempts to extract data from Alibaba's Amap, tried exposed API keys and probed pre-production servers, sometimes just to determine which zoo entrance people use.

A source close to Tencent says the agents were testing third-party models, not Hunyuan, and an internal review found no evidence of breaches.

@aitopics