Document Type

Presentation

Publication Date

7-2026

Keywords

artificial intelligence, AI privacy, data privacy, sensitive data, personally identifiable information, PII, large language models, generative AI, privacy-preserving AI, AI security, responsible AI, AI literacy

Disciplines

Education | Engineering

Abstract

This presentation examines privacy and data security risks associated with the growing use of artificial intelligence, with particular attention to the sensitive information users may disclose when interacting with AI systems. It explores how voice, text, images, and prompts can expose personally identifiable and privacy-sensitive information and introduces privacy threats associated with large language models and other AI systems. The presentation discusses approaches for identifying and protecting sensitive information, including named entity recognition, prompt modification, masking, replacement, and other privacy-preserving techniques. It also examines the trade-off between protecting privacy and maintaining the usefulness of AI-generated responses. Drawing on research in voice assistants, large language models, and vision-language models, the presentation demonstrates how seemingly routine interactions with AI can create privacy risks and highlights strategies for more privacy-conscious and responsible AI use.

Comments

Free and open access to this work is made available to the UMass Boston community by ScholarWorks at UMass Boston. Those not on campus and those without a UMass Boston campus username and password may gain access to this work through Interlibrary Loan. If you have a UMass Boston campus username and password and would like to download this work from off-campus, click on the “Off-Campus Users” button.

Community Engaged/Serving

Part of the UMass Boston Community-Engaged Teaching, Research, and Service Series. //scholarworks.umb.edu/engage

Rights

© 2026 Xiaohui Liang

Download

  Contact Author

Share

COinS