Remove PII from CSV Files with OpenAI

This workflow monitors a specified Google Drive folder for new CSV file uploads. Upon detection, it downloads the file, uses OpenAI to intelligently identify PII columns such as names, emails, phones, or addresses based on a customizable prompt, and filters them out to create a clean dataset.

The sanitized CSV is renamed (e.g., with '_cleaned' suffix) and uploaded to a separate Drive folder, preventing loops and ensuring originals remain intact. This is ideal for compliance with GDPR, CCPA, or HIPAA by automating data anonymization before sharing or processing.

Benefits include time savings on manual data scrubbing, reduced risk of data breaches, and scalability for high-volume uploads. Use cases: preprocessing customer data for analytics, preparing datasets for ML training, or routine privacy audits in marketing/sales pipelines.