Feature Request: Allow Users to Submit Product Feedback Directly Through ChatGPT
I would like to suggest a way for users to submit product feedback and feature requests to OpenAI directly from a ChatGPT conversation.
Currently, a user may be having a conversation with ChatGPT that naturally identifies a problem, missing feature, or potential improvement. By the time the discussion is finished, ChatGPT may already understand:
- What the user is trying to accomplish
- What limitation or friction the user encountered
- Why the existing behavior does not solve the problem
- What improvement the user is suggesting
- A real-world use case demonstrating why the feature would be useful
At that point, it seems unnecessary for the user to leave ChatGPT, locate a separate feedback channel, and reconstruct that entire explanation manually.
Proposed workflow
A user could simply say something such as:
“Submit this as feedback to OpenAI.”
ChatGPT could then:
- Recognize that the user wants to provide product feedback.
- Create a concise, structured feature request based on the relevant conversation.
- Show the proposed submission to the user before anything is sent.
- Allow the user to edit, approve, or cancel it.
- After explicit approval, submit the feedback through an internal OpenAI feedback system.
- Confirm that it was submitted and, if practical, provide a reference or tracking number.
The user should always be able to see exactly what information will be submitted before approving it.
This would not require exposing an email address
This does not necessarily need to be implemented as ChatGPT sending an email.
ChatGPT could simply have access to an internal product-feedback function or endpoint. The destination would not need to be publicly exposed, and users would not need to know or interact with the underlying system.
This could also allow OpenAI to receive feedback in a consistent, structured format instead of receiving thousands of differently formatted emails or forum posts.
Potential benefits to OpenAI
Because ChatGPT already understands the context that led to the suggestion, the resulting feedback could contain more useful information than a simple feature-request form.
The system could potentially categorize submissions by product area and identify substantially similar requests. This could help OpenAI understand how frequently users independently encounter the same problem without requiring every submission to be manually compared.
For example, instead of only receiving:
“Please add Feature X.”
OpenAI could receive structured information explaining:
Problem → Current workaround → Proposed improvement → Real-world use case
while still keeping the submission concise.
Privacy and user control
This should be explicitly opt-in.
ChatGPT should never automatically send a conversation to OpenAI as product feedback merely because it detects criticism or a suggestion.
Instead, the user initiates the process, ChatGPT prepares the submission, and the user approves exactly what will be shared.
Ideally, only the information necessary to explain the feedback would be included rather than automatically attaching an entire private conversation.
Why I am suggesting this
This feature request actually came about while discussing another feature request with ChatGPT.
During that conversation, ChatGPT and I developed the original idea, identified a real-world use case, refined how the feature could work, and considered privacy implications.
At that point, essentially all of the useful product feedback already existed inside the ChatGPT conversation.
The remaining process was to leave ChatGPT, find the appropriate feedback channel, create a forum topic, and manually transfer the information.
That seems like exactly the type of friction ChatGPT itself could eliminate.
In short:
If a user and ChatGPT have already developed useful product feedback together, the user should be able to review it, approve it, and send it to OpenAI without leaving the conversation.
