The community member is experiencing an error with the llama-parse library, where they are receiving an "UNKNOWN_ERROR: PDF_IS_BROKEN" message. They are unsure if the issue is on their end. Another community member has encountered a similar issue with 3 non-password protected PDFs, and suggests that if the original poster can share the files, it would be easier to help and potentially open a GitHub issue on the llama-parse repository.
Did you ever find out more about this? I have 3 non-password protected pdfs and it started happening to us today. I've transcribed these exact pdfs a few days ago