Job To Be Done:
Long-term users accumulate hundreds or thousands of conversations. Managing them individually is inefficient and doesn’t match real usage patterns — especially after Atlas increased the number of ultra-short chats (1–2 messages each).
Problem:
There is no way to clean up chat history in bulk.
Users can only delete conversations one by one, which becomes overwhelming as the number grows.
Short chats — accidental prompts, test queries, single-line checks — now dominate history.
Why it matters:
-
The current UX doesn’t scale with the new chat density.
-
Users lose control over their own data.
-
Large histories may increase storage load for OpenAI’s infrastructure, even if they don’t affect model inference.
-
Cleaning history manually is frustrating and discourages long-term use.
Request:
Add bulk-deletion tools that match real user behavior:
1. Delete All conversations
For users who want a clean slate or reset their workspace.
2. Delete Before Date
Example: “Delete everything before June 1, 2024.”
3. Delete Within a Date Range
Example: “Delete chats from Jan 1 to Mar 31.”
4. Delete All Chats With Fewer Than X Messages
Where X = 1–5 (user-selectable).
This is especially important after Atlas, where short “microprompts” generate dozens of tiny chats.
This feature would clean out all trivial one-off interactions without touching long, meaningful discussions.