Does anyone have a legitimate contact at OpenAI for product feedback and collaboration?

[BCBC]

Does anybody here have a connection at OpenAI—or know a public contact route that reaches someone with the authority to act on product feedback?

I have a proposal about making AI more useful for people building real projects: preserving work, maintaining continuity across sessions, making failures understandable, and connecting users with developers who can help turn their ideas into working tools.

This comes from extensive hands-on use, years of building communities, and decades working with software and infrastructure. I can explain the problems, show examples, and offer concrete ideas for improving the experience.

I’m looking for a legitimate introduction, a public business contact, or the right team to approach. Please don’t post anyone’s private contact details. An introduction with their permission would be appreciated.

I’m happy to send a short overview first so they can decide whether it belongs with their team.

Thanks, folks. Any help getting this in front of the right person would mean a lot.

[ai_daily_tools]

Honestly, the official developer forum is probably your best bet — OpenAI staff do read the feature request threads there. A detailed writeup like yours with concrete examples would do well on their community forum too. The hackathon and dev-day circuit is another route; people who present there tend to meet the partnership folks. Cold messages to product managers work sometimes, but a warm intro through someone who ships with their API carries way more weight.

[BCBC]

My experience was different. My post about regression—and the tool I built to track it—was deleted. Then I was asked why I was even there. I believe I have a screenshot, and I’ll check it before quoting anyone.

I was there to contribute. Reporting a problem and bringing a tool to help measure it should be the beginning of a useful conversation. Then this guy tell me I need to write a 500-1000 word essay about his site and why I am there??? LOL Sure FO Buddy!!

For some background: I came up with HeliFreak.com in a dream in 2003 after watching Henry “H to the C” Caldwell fly an R/C helicopter. I spent years working around the clock building that community before selling it in 2015. Other businesses I’ve been involved with include FreakTab.com, FlightPowerUSA.com, Computer Concepts, LTD, and Jamesnetworks.com, WealthCounsel.com and Interim Hellthcare I have Engineered/Designed/Implemented and written a few custom software packages for them.

I rarely bring that history up. I’m mentioning it because I have experience building businesses, listening to communities, and turning feedback into improvements. But nobody should need a résumé to have a reproducible problem taken seriously.

Here’s where I believe the priorities belong:

1. Improve ChatGPT itself. Reliability, consistency, continuity, and preserving the right project context. Stop making people repeat hours of work because

the system lost the decisions they already made.

2. Bring developers closer to the people using it. Make useful feedback welcome, give problems a clear route to investigation, and show people what was fixed and how it was verified.

Larger models can be valuable. So can fixing the everyday failures that waste people’s time. Leadership means deciding which improvement will actually help the person using the product.

I’m looking for a chance to contribute, with clear responsibilities, a modest salary, and results people can measure. I’ve helped change the direction of businesses before, and I believe those lessons apply here.

I don’t think AI’s future has to follow a self-defeating path. Its potential is enormous—but earning people’s trust takes listening, accountability, and follow-through.

Let’s change the direction together, one demonstrated improvement at a time.

[Grace2077]

I have had better luck with the official developer forum when the post is framed as a minimal reproducible case rather than a proposal. A short post with a paired example showing the regression, the date it changed, and the exact prompt used tends to survive moderation, while longer essays get moved or removed. The in-product feedback flow is another route that goes straight to the team and needs no forum approval. Keeping it to a few paragraphs with something they can rerun is what made the difference for me.

[BCBC]

Thank you for the info and experience.

The post was not a proposal at all just an observation I had not yet even created the logging tool. It is what it is.

They just keep ignoring the product that got them where they are… ChatGPT and the people who still use it… we will see how it goes.