PriView: A Human-AI Collaborative Tool for Facilitating Privacy Risk Review

Sharma, Tanusree

Permalink

https://hdl.handle.net/2142/121167 Copy

Description

Title

PriView: A Human-AI Collaborative Tool for Facilitating Privacy Risk Review

Author(s)

Sharma, Tanusree

Issue Date

2023-12-02

Keyword(s)

Privacy, Risk Assessment, Policy, Safety

Abstract

PriView tool aims to automate the evaluation of privacy requirements over time within evolving technologies and identify gaps in existing controls. It aims to bring organizations to a centralized privacy alignment across different teams and stakeholders by federating the process throughout the review. To achieve this, PriView creates a privacy taxonomy with manifested privacy risk scenarios relevant to various application domains (such as Cloud, Machine Learning, and IoT), as well as a model to assist organizations in privacy risk assessment. This taxonomy expands the ontology construction of risk-specific corpus, including information about how different applications operate, the types of data at risk, types of risk, and available baseline controls to mitigate risk. PriView also facilitates collaboration and education in making informed decisions for additional control recommendations. Using the established theoretical framework of Contextual Integrity, the tool illustrates how organizations can use the model to make informed decisions during privacy risk reviews, particularly concerning third-party applications. Finally, suggestions are provided for evaluating the model in practice with various stakeholders.

Type of Resource

text

Copyright and License Information

Tanusree Sharma

Owning Collections