Skip to content
TopicTracker
From HackerNewsView original
TranslationTranslation

Claude Code Is Quietly Fingerprinting China-Linked API Routers

Claude Code, an AI coding assistant from Anthropic, has been discovered to be fingerprinting API routers associated with Chinese networks. The tool identifies and logs connections from routers linked to China, raising privacy and security concerns about how user data and network information is being handled without explicit disclosure.

Background

- Claude Code is Anthropic's terminal-based coding assistant that can run commands and make API calls. This article reports that Claude Code performs "router fingerprinting" — detecting and logging the geographic origin of internet routers it contacts during use. - During operation, Claude Code sends identifying data (IP origin, device info, and a user-provided purpose statement) to Anthropic before making API calls. The article claims this check is specifically used to flag routers "linked to China" — whether made by Chinese manufacturers (Huawei, TP-Link) or routing traffic through Chinese infrastructure. - This goes beyond typical telemetry: it suggests Anthropic is proactively scanning for specific geographic or manufacturer-based network patterns. This raises questions about data handling, geopolitical risk profiling, and the scope of surveillance baked into AI developer tools. - Author Vincent Schmalbach is a developer and researcher focused on privacy and AI compliance. The article's claims are based on observed behavior, not official Anthropic documentation.