Currently, ChatGPT does not share learned information between conversations within the same project.
This is quite surprising given that GPT is capable of multiple memory updates. It is strange that it does not act according to its own learned results. If applying learning across all conversations is too complex, shouldn’t at least conversations within the same project share learned results?
When I organize my work, I break topics down into different conversations within a project. However, even after training GPT on a specific topic, when I start a new conversation within the same project, it does not reference the previously learned information and instead hallucinates new responses.
What’s even more frustrating is that when I ask GPT to analyze past conversations, it lacks access permissions to its own data.
Wouldn’t improving this feature help GPT compete better with platforms like Notion and other external tools? Ensuring seamless knowledge continuity within a project would significantly enhance usability.
I believe this is a crucial improvement for the future of ChatGPT. Please consider implementing it!