The Concept of Handoff as a Model for Ethical Analysis and Design
The Oxford Handbook of Ethics of AI Edited by Markus D. Dubber, Frank Pasquale, and Sunit Das
20 Pages Posted: 18 Feb 2021
Date Written: July 2020
This chapter introduces the concept of handoff, which offers a lens through which to evaluate sociotechnical systems in ethical and political terms. It is particularly tuned to transformations in which system components of one type replace components of another. Of great contemporary interest are handoff instances in which AI take over tasks previously performed by humans, for example, labelling images, processing and producing natural language, controlling other machines, predicting human action (and other events), and make decisions. Grounded in past work in social studies of technology and values in design, the handoff analytical model disrupts the idea that if components of a system are modular in functional terms, replacing one with another will leave ethical and political dimensions intact. Instead, the handoff lens highlights different ways that different types of system components operate and interoperate and shows these differences to be relevant to the configuration of values in respective systems. The handoff lens offers a means to make ethically relevant changes salient that might otherwise be overlooked.
Keywords: handoff analytical model, sociotechnical systems, automation, computational systems, human tasks, AI, ethical analysis
Suggested Citation: Suggested Citation