I had to start a new chat because the responses were acting weird, but now it doesn’t recognize the PDF that worked fine a day ago. it keeps saying things ‘It seems you uploaded some files, how would you like me to process them’, ‘It looks like you’ve uploaded some files, but I can’t see what’s in them yet. What are you trying to find or analyze?’ When I ask it to analyze, summarize, or extract text, it responds with ‘(Summarizing…), (Reading the file…), (Inserting the last paragraph…), [Inserts the actual summary of the last paragraph]’ as if it’s a role-playing game. It would be great if this could be fixed
3 Likes
same. anyone one else getting this?
2 Likes
yup hates working with documents now its referencing stuff that has nothing to do with the document i provided as well.
2 Likes
Yeh I just noticed this shit today when i’m trying to use it to help me out on a couple exam questions as I do normally but today it keeps on repeating “It looks like you’ve uploaded some files. How can I help you with them? Are you looking for specific information or analysis?” and other shit too. This was not happening before today at all.
2 Likes
Should be same as my issue. 4-turbo is being used instead of 4o, and it is hallucinating.
Try contacting support directly as well
Same. Has anyone been able to get more info on this? It started several days ago for me and still not working correctly.
2 Likes
me too. please someone help with a fix, it cant read any files at all recently.
2 Likes
Not sure, but it is worse in projects than in normal chat thread. At least in normal chat thread it doesn’t hallucinate. In projects, it is hallucinating.
In normal chat thread it has a clear cutoff of 10000~12000 words. ChatGPT will read until around 12000, but doesn’t hallucinate. They just don’t have information after that point.
In Projects, it is not sure where it starts hallucinating
I have managed to salvage somethings from my project. Basically, once your files goes above 12000 words in projects, chatgpt will hallucinate about that file, they will only read about 2000 words, then fill in the rest with hallucination. Other files in the same projects are able to be read properly(As long as they don’t need to be integrated to 1 file).
If you keep your files under 12000 words, you might be able to do work
I haven’t seen this news on social media. All developers should really retest their chatgpt projects/custom gpts for breakage.
You should all tell your friends in case theirs broke and they don’t know. Multimillion dollars deals or other critical situations might break.