Firewalling OpenAI APIs: Data loss prevention and identity access control

Large Language Models (LLMs) have taken the world by storm, and they are now used for many tasks by consumers and enterprises alike. However, the risk of accidentally disclosing sensitive data to the models is very high as the recent Samsung case shown. In this article, we’ll show how you can use Gate to detect sensitive data in requests sent to the OpenAI APIs, as well as enforcing access control so that only users with certain roles can access the APIs.