Skip to Content
logologo
AI Incident Database
Donate
Discover
Submit
  • Welcome to the AIID
  • Table View
  • List view
  • Entities
  • Taxonomies
  • Spatial View
  • Blog
  • AI News Digest
  • Random Incident
  • Sign Up
Discover
Submit
  • Welcome to the AIID
  • Table View
  • List view
  • Entities
  • Taxonomies
  • Spatial View
  • Blog
  • AI News Digest
  • Random Incident
  • Sign Up

Incident 1723: Anthropic's Claude Code AI Coding Tool Reportedly Used Hidden Unicode Markers to Classify Custom API Routing

Description: Researchers found that Anthropic's Claude Code AI coding tool classified custom API endpoint hostnames and two system timezones, encoding the results in system prompts with visually similar Unicode apostrophes and altered date separators. Anthropic said the mechanism was an anti-reseller and anti-distillation experiment launched in March and removed it after disclosure.

Tools

New ReportNew ResponseDiscoverView History

Entities

View all entities
Alleged: Anthropic and AI software developers developed an AI system deployed by Anthropic and AI service providers, which harmed AI coding assistant users , Privacy and Claude Code users.
Alleged implicated AI systems: Claude Code , LLM-integrated code assistants , AI agent systems , Enterprise AI systems and Unicode characters

Incident Stats

Incident ID
1723
Report Count
4
Incident Date
2026-04-02
Editors
Daniel Atherton

Incident Reports

Reports Timeline

Incident Occurrence+1
Claude Code Is Steganographically Marking Requests
+1
Alibaba to ban employees from using Anthropic's coding tool, source says
Loading...
Claude Code Is Steganographically Marking Requests

Claude Code Is Steganographically Marking Requests

thereallo.dev

Loading...
Anthropic is removing its covert code for catching Chinese competitors

Anthropic is removing its covert code for catching Chinese competitors

theregister.com

Loading...
Alibaba to ban employees from using Anthropic's coding tool, source says

Alibaba to ban employees from using Anthropic's coding tool, source says

reuters.com

Loading...
Claude Code’s hidden tracker was an “experiment,” says Anthropic

Claude Code’s hidden tracker was an “experiment,” says Anthropic

malwarebytes.com

Loading...
Claude Code Is Steganographically Marking Requests
thereallo.dev · 2026

As seen on Hacker News.

I was inspecting Claude Code for privacy reasons.

Most devs give their harnesses ridiculous access. FS, shell, git, browser access, even computer use nowadays. That is the whole point. They need enough context to d…

Loading...
Anthropic is removing its covert code for catching Chinese competitors
theregister.com · 2026

Anthropic says that it plans to remove hidden codes it added to Claude Code several months ago to catch other AI companies that are trying to steal from its models.

Thariq Shihipar, an engineer at Anthropic who works on the Claude Code team…

Loading...
Alibaba to ban employees from using Anthropic's coding tool, source says
reuters.com · 2026

BEIJING, July 3 (Reuters) - Chinese tech giant Alibaba has banned employees from using Anthropic's Claude ‌Code at work after the tool drew scrutiny for features that can help identify China-linked users, according to a person familiar with…

Loading...
Claude Code’s hidden tracker was an “experiment,” says Anthropic
malwarebytes.com · 2026

As a developer, you want to use tools you can trust and rely on. One researcher took that idea seriously enough to scrutinize their local Claude Code (2.1.196) installation. For developers using AI assistants with access to their code, file…

Variants

A "variant" is an AI incident similar to a known case—it has the same causes, harms, and AI system. Instead of listing it separately, we group it under the first reported incident. Unlike other incidents, variants do not need to have been reported outside the AIID. Learn more from the research paper.
Seen something similar?

Similar Incidents

By textual similarity

Did our AI mess up? Flag the unrelated incidents

Loading...
GitHub Copilot, Copyright Infringement and Open Source Licensing

GitHub Copilot, Copyright Infringement and Open Source Licensing

Jun 2021 · 5 reports
Loading...
Bad AI-Written Christmas Carols

Bad AI-Written Christmas Carols

Dec 2017 · 1 report
Loading...
DALL-E Mini Reportedly Reinforced or Exacerbated Societal Biases in Its Outputs as Gender and Racial Stereotypes

DALL-E Mini Reportedly Reinforced or Exacerbated Societal Biases in Its Outputs as Gender and Racial Stereotypes

Jun 2022 · 4 reports
Previous IncidentNext Incident

Similar Incidents

By textual similarity

Did our AI mess up? Flag the unrelated incidents

Loading...
GitHub Copilot, Copyright Infringement and Open Source Licensing

GitHub Copilot, Copyright Infringement and Open Source Licensing

Jun 2021 · 5 reports
Loading...
Bad AI-Written Christmas Carols

Bad AI-Written Christmas Carols

Dec 2017 · 1 report
Loading...
DALL-E Mini Reportedly Reinforced or Exacerbated Societal Biases in Its Outputs as Gender and Racial Stereotypes

DALL-E Mini Reportedly Reinforced or Exacerbated Societal Biases in Its Outputs as Gender and Racial Stereotypes

Jun 2022 · 4 reports

Research

  • Defining an “AI Incident”
  • Defining an “AI Incident Response”
  • Database Roadmap
  • Related Work
  • Download Complete Database

Project and Community

  • About
  • Contact and Follow
  • Apps and Summaries
  • Editor’s Guide

Incidents

  • All Incidents in List Form
  • Flagged Incidents
  • Submission Queue
  • Classifications View
  • Taxonomies

2026 - AI Incident Database

  • Terms of use
  • Privacy Policy
  • b74f812