[Feature Request] Workspace-Level Privacy Settings for Training Consent

[ewheelerinc]

Bug Description Unfortunately, I have to set my retraining-setting to private (30-day retention) because a very small part of my job deals with private information.

However, most of my work is on open-source software. This means that Claude does not benefit from learning and training based on my normal everyday duties. I think it would be useful if the Claude code software allowed you to select the privacy setting by workspace and choose an appropriate default.

When a new workspace is first loaded, it would prompt for the privacy setting, probably defaulting to "allow training"

This would allow users to control their privacy while supporting training of programming use patterns. It would also align with the project-selected opt-in training settings that Open AI has, and would put you at an additional competitive advantage by making this easy.

Environment Info

• Platform: linux
• Terminal: vte-based
• Version: 2.0.8
• Feedback ID:

Errors

[]

[github-actions[bot]]

This issue has been inactive for 30 days. If the issue is still occurring, please comment to let us know. Otherwise, this issue will be automatically closed in 30 days for housekeeping purposes.

[bionicles]

Privacy concerns persist, please keep this issue open