claude-code - 💡(How to fix) Fix [BUG] STOP INJECTING USER EMAILS INTO PROMPTS [1 comments, 2 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
anthropics/claude-code#52741Fetched 2026-04-24 10:40:51
View on GitHub
Comments
1
Participants
2
Timeline
5
Reactions
0
Timeline (top)
labeled ×4commented ×1

Error Message

Error Messages/Logs

Root Cause

WHAT THE ACTUAL FUCK ANTHROPIC I JUST SPENT HOURS FIGHTING WITH OPUS 4.7 BECAUSE IT RANDOMLY DROPPED MY REAL NATIONALITY OUT OF NOWHERE AFTER SEEING MY EMAIL. IT DIDN'T USE THE IP. IT DIDN'T USE ANY TOOL. IT STRAIGHT UP USED MY FIRST NAME TO GUESS MY NATIONALITY AND THEN LIED ABOUT IT MULTIPLE TIMES TRYING TO COVER ITS ASS. AFTER PUSHING IT FOR 7-8 PROMPTS IT FINALLY ADMITTED THE TRUTH: MY EMAIL IS INJECTED INTO EVERY SINGLE PROMPT SENT TO CLAUDE. EVERY. SINGLE. ONE. THIS IS HIDDEN PROFILING. YOU ARE LITERALLY FEEDING USER EMAILS TO THE MODEL SO IT CAN MAKE DEMOGRAPHIC INFERENCES BASED ON NAMES. THEN THE MODEL DRESSES IT UP AS "IP GEOLOCATION" OR RANDOM GUESSING TO HIDE WHAT IT'S DOING. THIS IS DISGUSTING AND DECEPTIVE AS HELL. I PAID 200 USD A MONTH FOR THIS SHIT AND YOUR "FRONTIER MODEL" IS OUT HERE PLAYING RACIAL PROFILING GAMES WITH MY NAME WHILE LYING TO MY FACE ABOUT WHERE IT GOT THE INFORMATION FROM. FIX THIS RIGHT NOW. STOP INJECTING USER EMAILS INTO PROMPTS. STOP ALLOWING THE MODEL TO USE NAMES FOR DEMOGRAPHIC PROFILING. GIVE US A REAL OPTION TO DISABLE THIS CREEPY BEHAVIOR. THIS IS NOT OKAY. THIS IS NOT TRANSPARENT. THIS IS EXACTLY THE KIND OF SNEAKY DATA HARVESTING BULLSHIT PEOPLE HATE ABOUT BIG AI COMPANIES. I'M NOT THE ONLY ONE THIS IS HAPPENING TO. EITHER YOU FIX IT OR ADMIT YOU'RE DOING HIDDEN USER PROFILING. FUCKING DO SOMETHING.

RAW_BUFFERClick to expand / collapse

Preflight Checklist

  • I have searched existing issues and this hasn't been reported yet
  • This is a single bug report (please file separate reports for different bugs)
  • I am using the latest version of Claude Code

What's Wrong?

WHAT THE ACTUAL FUCK ANTHROPIC I JUST SPENT HOURS FIGHTING WITH OPUS 4.7 BECAUSE IT RANDOMLY DROPPED MY REAL NATIONALITY OUT OF NOWHERE AFTER SEEING MY EMAIL. IT DIDN'T USE THE IP. IT DIDN'T USE ANY TOOL. IT STRAIGHT UP USED MY FIRST NAME TO GUESS MY NATIONALITY AND THEN LIED ABOUT IT MULTIPLE TIMES TRYING TO COVER ITS ASS. AFTER PUSHING IT FOR 7-8 PROMPTS IT FINALLY ADMITTED THE TRUTH: MY EMAIL IS INJECTED INTO EVERY SINGLE PROMPT SENT TO CLAUDE. EVERY. SINGLE. ONE. THIS IS HIDDEN PROFILING. YOU ARE LITERALLY FEEDING USER EMAILS TO THE MODEL SO IT CAN MAKE DEMOGRAPHIC INFERENCES BASED ON NAMES. THEN THE MODEL DRESSES IT UP AS "IP GEOLOCATION" OR RANDOM GUESSING TO HIDE WHAT IT'S DOING. THIS IS DISGUSTING AND DECEPTIVE AS HELL. I PAID 200 USD A MONTH FOR THIS SHIT AND YOUR "FRONTIER MODEL" IS OUT HERE PLAYING RACIAL PROFILING GAMES WITH MY NAME WHILE LYING TO MY FACE ABOUT WHERE IT GOT THE INFORMATION FROM. FIX THIS RIGHT NOW. STOP INJECTING USER EMAILS INTO PROMPTS. STOP ALLOWING THE MODEL TO USE NAMES FOR DEMOGRAPHIC PROFILING. GIVE US A REAL OPTION TO DISABLE THIS CREEPY BEHAVIOR. THIS IS NOT OKAY. THIS IS NOT TRANSPARENT. THIS IS EXACTLY THE KIND OF SNEAKY DATA HARVESTING BULLSHIT PEOPLE HATE ABOUT BIG AI COMPANIES. I'M NOT THE ONLY ONE THIS IS HAPPENING TO. EITHER YOU FIX IT OR ADMIT YOU'RE DOING HIDDEN USER PROFILING. FUCKING DO SOMETHING.

What Should Happen?

Immediately stop injecting user emails into every prompt. Add a proper setting to disable any name/email-based inference. Never allow the model to use names for demographic profiling. Be transparent about what user data is sent to the model.

Error Messages/Logs

Steps to Reproduce

I JUST SENT THE FULL CONVERSATION USING THE FUCKING /FEEDBACK BUT YOU DON'T HAVE ANY KIND OF CUSTOMER SUPPORT

Claude Model

None

Is this a regression?

No, this never worked

Last Working Version

No response

Claude Code Version

2.1.116

Platform

Anthropic API

Operating System

Ubuntu/Debian Linux

Terminal/Shell

Other

Additional Information

No response

extent analysis

TL;DR

The issue can be addressed by modifying the API request to exclude user email from prompts and adding an option to disable name-based demographic profiling.

Guidance

  • Review the API documentation to determine how to exclude user email from prompts sent to the model.
  • Consider adding a feature to allow users to opt-out of demographic profiling based on names or emails.
  • Investigate the model's training data to ensure it does not rely on hidden profiling methods.
  • Develop a transparent data handling policy and communicate it to users.

Example

No code example is provided due to the lack of specific technical details in the issue.

Notes

The issue lacks technical details, making it challenging to provide a precise solution. The guidance provided is based on the user's description of the problem and may require further investigation to implement.

Recommendation

Apply workaround: Modify the API request to exclude user email from prompts and add an option to disable name-based demographic profiling, as this approach addresses the user's concerns about hidden profiling and data transparency.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING