Privacy-Enhanced AI Assistants Based on Dialogues and Case Similarity

Abstract

Personal assistants (PAs) such as Amazon Alexa, Google Assistant and Apple Siri are now widespread. However, without adequate safeguards and controls their use may lead to privacy risks and violations. In this paper, we propose a model for privacy-enhancing PAs. The model is an interpretable AI architecture that combines 1) a dialogue mechanism for understanding the user and getting online feedback from them, with 2) a decision-making mechanism based on case-based reasoning considering both user and scenario similarity. We evaluate our model using real data about users’ privacy preferences, and compare its accuracy and demand for user involvement with both online machine learning and other, more interpretable, AI approaches. Our results show that our proposed architecture is more accurate and requires less intervention from the users than existing approaches.

Healthcare Voice AI Assistants: Factors Influencing Trust and Intention to Use

“…Researchers are actively working on a range of solutions, such as mechanisms for providing user with meaningful verbal consent [31,162] and automated systems to identify policy breaches [52,207]. Additionally, significant efforts are being directed towards creating models that aid users in making privacy decisions consistent with their preferences [208,209] and crafting explanations that incorporate elements of trust and privacy, which are intended to diminish user concerns and strengthen trust in VAs [163,174].Moreover, how users' perceive security and privacy of these underlying platforms is very important for HVAs, as users consider healthcare-related data more sensitive and prefer to share less information with voice assistants about health [3]. As the information they share with them is often protected by additional regulation (such as HIPAA in the U.S.), the security of those HVAs is therefore highly dependent on the security of the voice assistants on top of which they are built.…”