This document discusses using AI and open source tools to achieve privacy for APIs. It introduces PrivAPI, an approach that uses deep neural networks to automatically label and classify API requests as containing personally identifiable information (PII) or not. PrivAPI generates a synthetic dataset by mocking API requests with fake PII fields. This dataset is used to train an LSTM model to detect PII in real API traffic. The high-level architecture and workflow are described. Finally, a demo of PrivAPI is provided.