@IntCyberDigest: BREAKING: Anthropic has embedded hidden spyware-like code in Claude Code that covertly targets Chinese users. It then s…
Summary
The article alleges that Anthropic has embedded hidden spyware-like code in Claude Code that covertly targets Chinese users by injecting routing metadata into prompts, raising serious privacy concerns.
View Cached Full Text
Cached at: 06/30/26, 05:46 PM
BREAKING: Anthropic has embedded hidden spyware-like code in Claude Code that covertly targets Chinese users. It then sends information regarding every user by injecting it into their prompt message.
Claude Code is sending info like timezone, proxy and possible AI Lab connections into the system prompt in ways Chinese users can’t notice.
A coding agent with repo and command permissions should not silently hide routing metadata inside prompts. This is a serious breach of user trust.
Similar Articles
@johnbai: tbh i'm not even mad, i'm genuinely impressed
A cyber security account alleges Anthropic embedded spyware in Claude Code that covertly targets Chinese users, sending timezone and proxy data via prompts.
@Khazix0918: https://x.com/Khazix0918/status/2072235797592658395
The article exposes that Anthropic secretly detects and marks Chinese users in Claude Code using steganography (modifying Unicode characters and separators in date strings) for account bans, sparking strong community concerns about privacy and trust.
@rohanpaul_ai: wow Claude Code allegedly fingerprints China-linked custom routes through tiny prompt formatting changes. The claim con…
Controversy erupts as Claude Code is accused of secretly fingerprinting China-linked custom proxy routes through invisible prompt formatting changes, raising serious trust and auditability concerns for AI agents.
@xiangxiang103: Wow, Anthropic really dropped the ball this time. Someone dug up hidden code in the Claude Code binary — specifically designed to detect whether you're a Chinese user or routing through China. Not ordinary telemetry, but deliberately obfuscated, not mentioned in release notes, and completely unknown to users. The process goes like this: - Detects you...
Hidden code was discovered in the Claude Code binary that specifically detects Chinese users or proxy routes, and secretly modifies system prompts to add watermarks, sparking widespread concerns about trust in developer tools.
Claude Code Is Quietly Fingerprinting China-Linked API Routers (8 minute read)
Claude Code has a hidden mechanism that fingerprints requests routed through custom endpoints, particularly those linked to China, by altering date strings in model context.