Out of Context: Investigating the Bias and Fairness Concerns of “Artificial Intelligence as a Service”

By Kornel Lewicki et al.

Published on Jan. 1, 2023

Read the original document by opening this link in a new tab.

Table of Contents

1. ABSTRACT
2. INTRODUCTION
3. BACKGROUND
4. AI as a Service (AIaaS)
5. AI Regulation
6. Algorithmic Fairness
7. Fairness and AIaaS

Summary

The paper discusses the bias and fairness concerns associated with the growing market of 'AI as a Service' (AIaaS). It highlights the challenges of replicating biases in AI systems and the contextual nature of fairness in the AIaaS paradigm. The authors propose a taxonomy of AIaaS offerings based on levels of autonomy afforded to users and critically examine the fairness concerns and practical challenges of different AIaaS categories. The paper aims to draw attention to the implications of AIaaS models and encourages discussions on responsible development and use of AI services.