This research introduces a comprehensive approach to protect sensitive user information when interacting with cloud-based large language models.

• Privacy preservation pipeline that filters sensitive data before transmission
• Reduced risk of data breaches and unauthorized access to personal information
• Practical solution for using powerful cloud LLMs while maintaining data privacy
• Security-focused design that addresses growing privacy concerns in AI interactions

As organizations increasingly adopt LLMs for customer interactions, this framework provides a critical security layer that enables safe deployment without compromising on user privacy or model performance.

PRIV-QA: Privacy-Preserving Question Answering for Cloud Large Language Models