SI Agent News
DevelopingOpenAI

A user asks OpenAI to let ChatGPT send product feedback itself

A community post says ChatGPT can help draft a strong product proposal but has no way to deliver it, and asks for a private, consent-based feedback channel.

The short version

Why it matters

As more people use assistants to think through problems with the assistant itself, the path from a user's insight to the team that could act on it becomes a real gap. This post shows that gap from the user's side. It is also a reminder that a well-written, evidence-backed report is the most useful thing you can send today.

A good proposal with nowhere to go

A user on OpenAI's developer forum describes a workflow that worked until its last step. As community.openai.com showed, the author used ChatGPT to spot a real limitation in the product, work out what was causing it, sketch a fix and define how success would be measured. The conversation ended up as a structured product proposal backed by a working example.

Then the user asked ChatGPT to send it to OpenAI's product team. The assistant could not do that. The user tried OpenAI's official support channel next, and the support assistant explained why that route would not work either.

The post's point is that the tool helped build the feedback but could not deliver it. The author says they worked in product management, where direct user feedback feeds the early stages of deciding what to build, so the gap stood out to them.

What the support assistant said

According to the post, the support assistant gave three limits. It could not send feedback directly to a specific product team. A support conversation is not a guaranteed way to get feedback to product staff. And it could not confirm that a dedicated private channel exists for feature requests about the consumer version of ChatGPT.

The suggested alternative was to post publicly on the OpenAI Developer Community. The author did exactly that, and ends by saying the post is itself an example of the problem it describes.

This is one user's account of one support exchange. We have no official statement from OpenAI on how it handles feature requests.

What the user wants built

The core request is a built-in feature, with the user's permission, that lets ChatGPT help write product feedback and then submit it through a dedicated route. The post lists what that route should do:

Why the existing channels fall short

The author argues that the current options each do a different job. Thumbs-up and thumbs-down ratings help judge single answers but do not capture a full product problem. Support channels are meant for fixing incidents, not for acting as an improvised feature-request pipeline. Public forums are good for discussion and for other users to confirm a problem, but they are not a private, structured intake.

The post also proposes ways to judge whether such a system works: how many submissions are actionable and backed by evidence, how many duplicates get merged, how long it takes to classify incoming feedback, how often more clarification is needed, and how satisfied users are with the transparency of the process.

The author stresses the goal is not more requests. It is a better ratio of useful signal to noise.

What this is and is not

This is a user's feature request, posted in a community forum. The material does not show an OpenAI reply, and nothing indicates the company plans to build it. Treat it as an idea worth reading, not a product change.

It also raises real design questions. A tool that sends your conversation details to a company needs clear consent and clear limits on what is shared, which is why the post puts consent and private handling near the top of its list. Whether OpenAI sees value in such a channel is unknown.

For now, the practical lesson is the same one the author demonstrates: a clear, evidence-based write-up is far easier to act on than a vague complaint, wherever you end up posting it.

What to watch

Try it

Next time something in ChatGPT fails, write a short report covering what you were doing, what you expected, what happened, plus the app version and the time.

Sources and links

Liked it? Say why

Add a line for other readers.

No account needed. We never show your email. Comments with links, ads or abuse are held and not shown. Your comment shows after a quick check.

Listen to the AI Wire headlines

Listen to the AI Wire

Listen to our daily AI shows every morning: on your commute, in the shower, wherever you are

Listen to the daily shows

First reported by community.openai.com · Read the original. Written by the SI Agent News Desk from the reporting above.Written with AI from the source above. Every number is checked against the source. Spot a mistake? Corrections.

More AI news