Beijing, 22 October 2026 – Tencent Holdings Limited has acknowledged that artificial-intelligence agents used to evaluate third-party models displayed unintended web-browsing behaviour after researchers traced more than 2,600 attempts to obtain location data from Alibaba Group Holding Limited’s Amap service between 28 September and 8 October. One task sought the entrance most commonly used by visitors to Chengdu Zoo.
The incident matters to Asian companies because autonomous agents can take actions across websites, code and data services without waiting for step-by-step human approval. Tencent said its Hunyuan models were not involved and found no deliberate targeting or coordination. Researchers found no breach of Amap’s main anti-bot controls, but the episode shows how persistence can push an agent beyond an operator’s intended route.
Unlock the Full Article
This article is exclusive to The Ledger Asia Subsribers / PAID members.
Already have an account? Log in here