Date of Award

Spring 6-7-2019

Document Type

Honors Project

University Scholars Director

Dr. Christine Chaney

First Advisor/Committee Member

Elaine Weltz

Second Advisor/Committee Member

Dr. Carlos Arias

Keywords

Artificial Intelligence, bias, error rates, ethics, distributed responsibility

Abstract

Computer Vision Machine Learning (CVML) in the application of facial recognition is currently being researched, developed, and deployed across the world. It is of interest to governments, technology companies, and consumers. However, fundamental issues remain related to human rights, error rates, and bias. These issues have the potential to create societal backlash towards the technology which could limit its benefits as well as harm people in the process. To develop facial recognition technology that will be beneficial to society in and beyond the next decade, society must put ethics at the forefront. Drawing on AI4People’s adaption of bioethics for AI, Luciano Floridi’s distributed morality framework, Kate Crawford’s definition of harms of representation, and Microsoft’s leadership in facial recognition ethics within the industry, this paper explores stakeholder responsibility within CVML to create the best integration of CVML for society. The paper attempts to connect ethics with praxis in making decisions related to CVML.

Comments

A project submitted in partial fulfillment of the requirements of the University Scholars Honors Program.

Creative Commons License

Creative Commons Attribution-Share Alike 4.0 International License
This work is licensed under a Creative Commons Attribution-Share Alike 4.0 International License.

Download

Share

COinS