I have a Plus subscription, and I understand that I should have access to the feature of uploading documents for analysis with ChatGPT. However, when I try to upload a document for analysis, ChatGPT doesn’t seem to be able to see or process it.
I have followed the necessary steps for uploading the document, but every time I try to interact with it, I do not receive a corresponding response. Could you please help me understand if there is a technical issue or a setting I need to enable in order to use this feature?
There seems to be a empty market design niche between API based and obscure layer web interface users of the same models.
We do not have awareness or access as prompters to something called system prompt or anything about conversation initialization, that would allow us to understand or compensate how the model is handling the most recent models not having been in their training corpus.
So, I have had to waste (or spend, or work ) as a reverse problem, hitting inconstistencies of behavior because of this overlooked problem of conversation operational data that neither user or assistant seem to have.
I have tried imposing the data that we can find in the API documentation for the same model (4o) for example… I am now at a point, of trying the right size of input that should work. but given the many past disappointments contrasting with the inferred belief given the advertised claims, and the past conversation long discussions with the model planning again and again for a more testable and successful long document context project, I fear putting my hand in such desparing adventure… I have done all the various volubile chat bots, including o1 for their reasoning about such plans for putative other models (target being 4o). But I would not say I am confident it will work.
it might be another BS adventure of a few months. I learned a lot. but that was not my project.
It’s sad, but it’s happening.