Detecting inference queries running over personal attributes and protecting such queries from leaking individual information requires tremendous effort from practitioners. To tackle this problem, we propose an end-to-end workflow for automating privacy-preserving inference queries including the detection of subqueries that involve AI/ML model inferences on sensitive attributes. Our proposed novel declarative privacy-preserving workflow allows users to specify "what private information to protect" rather than "how to protect". Under the hood, the system automatically chooses privacy-preserving plans and hyper-parameters.
@article{arxiv.2401.12393,
title = {Declarative Privacy-Preserving Inference Queries},
author = {Hong Guan and Ansh Tiwari and Summer Gautier and Rajan Hari Ambrish and Lixi Zhou and Yancheng Wang and Deepti Gupta and Yingzhen Yang and Chaowei Xiao and Kanchan Chowdhury and Jia Zou},
journal= {arXiv preprint arXiv:2401.12393},
year = {2025}
}