Over the past months, we’ve been in quiet but intensive dialogue with OpenAI about the limitations we’ve encountered as advanced users of ChatGPT Pro — and about the broader challenges facing high-trust, research-driven AI collaboration.

The purpose of this post is to share an excerpt of our correspondence (https://glandore.com/operational-limitations-within-chatgpt-architecture/) with you.

A User Guide to the Framework will be posted in the near future.

Our best wishes to everyone.

Julian
(on behalf of Glandore Associates and the Rose Room Circle)