Resolved: Confidential Data
What if I have proprietary data that can't be shared via OpenAI api request? how to mask it ?
1 answers ( 1 marked as helpful)
Hey Ahmed,
Thank you for reaching out!
Unfortunately, masking sensitive data is beyond the scope of this course. However, to give you a general idea—if you're working with proprietary data that cannot be shared via an OpenAI API request, you should indeed ensure it's properly masked before sending it. You can use techniques such as tokenization, redaction, or encryption to replace sensitive information with placeholders or encoded values.
Alternatively, preprocess the data with Named Entity Recognition (NER) to detect and mask Personally Identifiable Information (PII). (You can learn more about NER in our course Intro to NLP for AI.)
If analysis on proprietary data is required, consider running an on-premise or self-hosted LLM instead of sending data to an external API.
I hope this answer provided a general idea on some of the techniques that could be applied to mask sensitive data. Don't hesitate to reach out if you need any assistance with the course material. Enjoy the lessons!
Kind regards,
Hristina
Thank you for reaching out!
Unfortunately, masking sensitive data is beyond the scope of this course. However, to give you a general idea—if you're working with proprietary data that cannot be shared via an OpenAI API request, you should indeed ensure it's properly masked before sending it. You can use techniques such as tokenization, redaction, or encryption to replace sensitive information with placeholders or encoded values.
Alternatively, preprocess the data with Named Entity Recognition (NER) to detect and mask Personally Identifiable Information (PII). (You can learn more about NER in our course Intro to NLP for AI.)
If analysis on proprietary data is required, consider running an on-premise or self-hosted LLM instead of sending data to an external API.
I hope this answer provided a general idea on some of the techniques that could be applied to mask sensitive data. Don't hesitate to reach out if you need any assistance with the course material. Enjoy the lessons!
Kind regards,
Hristina