Summary
Please add duplicate-file detection, comparison, and bulk-cleanup tools to ChatGPT Library.
As users upload more documents across different conversations and projects, the Library can quickly accumulate multiple copies of the same file. At present, there is no practical way to identify exact duplicates, compare similar versions, or remove duplicate files in bulk.
Current problem
Users may unknowingly upload the same file multiple times because:
-
The same document is used in different conversations.
-
Files are renamed before being uploaded.
-
Updated and older versions have similar names.
-
It is difficult to confirm whether a file already exists in the Library.
-
Deleting a conversation does not necessarily provide a clear way to manage the corresponding Library files.
Manual cleanup becomes inefficient when the Library contains hundreds or thousands of files.
Suggested functionality
1. Exact duplicate detection
Detect byte-for-byte identical files using SHA-256 or another reliable cryptographic hash.
The Library should group files with identical contents even when:
-
Their filenames are different.
-
They were uploaded at different times.
-
They came from different conversations or projects.
2. Similar-document and version detection
Identify files that are not binary-identical but contain substantially the same content, such as:
-
Different versions of the same Word document.
-
PDFs generated from the same source.
-
Files with minor formatting or metadata changes.
-
Documents with small textual revisions.
Users should be able to compare the differences before deciding which version to keep.
3. Duplicate groups
Display suspected duplicates in clear groups, including:
-
Filename
-
File type
-
File size
-
Upload date
-
Source conversation or project
-
Exact-match or similarity status
-
File hash for exact duplicates
-
Version or content differences where available
4. Safe bulk cleanup
Allow users to:
-
Keep the newest file
-
Keep the oldest file
-
Keep the file linked to the most conversations
-
Select one file manually
-
Delete all remaining duplicates in the group
No files should be deleted automatically without explicit user confirmation.
5. Review report and recovery
Before deletion, generate a review report showing:
-
Files that will be retained
-
Files that will be deleted
-
Estimated storage space recovered
-
Conversations or projects that may reference each file
Deleted files should remain recoverable through Recently Deleted for an appropriate period.
6. Chat-based Library analysis
Allow users to ask ChatGPT questions such as:
-
“Find duplicate files in my Library.”
-
“Show me files with identical contents.”
-
“Compare different versions of this document.”
-
“Create a safe cleanup plan without deleting anything.”
-
“Show me duplicates larger than 50 MB.”
ChatGPT should initially provide an analysis and proposed action list. Deletion should require a separate confirmation step.
7. Prevent future duplicates
When a user uploads a file that already exists in the Library, ChatGPT could display a notice such as:
An identical file already exists in your Library. Would you like to reuse the existing file instead of uploading another copy?
This could reduce unnecessary storage usage before duplicates are created.
Privacy and safety considerations
-
Hash calculations should be performed securely.
-
Content comparison should respect existing privacy and workspace permissions.
-
Users should be able to disable similarity analysis.
-
Files from different workspaces or organizations should never be compared unless permitted.
-
Destructive actions should always require explicit confirmation.
-
A cleanup operation should produce an audit log or downloadable report.
Expected benefit
This feature would:
-
Make the Library easier to organize
-
Reduce unnecessary storage usage
-
Prevent accidental deletion of important versions
-
Improve document reuse across conversations
-
Make ChatGPT more effective for users who manage large numbers of files
A Library intended for long-term file reuse should include native tools for duplicate detection, version comparison, and safe bulk cleanup.