Hello and thank you for your time, I hope this message finds you well.
I am currently utilizing the GPT-4 Image-to-Text API for processing images of identification documents such as IDs and passports.
Until recently, the API provided satisfactory OCR results, which we integrated into our database system for various applications.
As of today, I’ve observed that the assistant is declining to process these images, indicating an inability to assist with personal data processing.
Can we please clarify if this change is expected from OpenAI’s side?
Has there been an update to the API or its usage policies that restricts handling personal identification documents?
I’m aware that there are other OCR solutions available that offer accuracy, reliability, and security.
However, integrating GPT-4’s capabilities was beneficial for our use case.
Understanding whether OCR for personal documents is no longer supported will help us adjust our implementation accordingly.
Thank you for once again for your time and assistance!