Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Data Ethics Emergency Drill: A Toolbox for Discussing Responsible AI for Industry Teams

Vanessa Aisyahsari Hanschke, Dylan Rees, Merve Alanyali, David Hopkinson, Paul Marshall

TL;DR

The findings show that practitioners can apply lessons learnt from the roleplay to real-life situations, and how the DEED opened up conversations around ethics and values.

Abstract

Researchers urge technology practitioners such as data scientists to consider the impacts and ethical implications of algorithmic decisions. However, unlike programming, statistics, and data management, discussion of ethical implications is rarely included in standard data science training. To begin to address this gap, we designed and tested a toolbox called the data ethics emergency drill (DEED) to help data science teams discuss and reflect on the ethical implications of their work. The DEED is a roleplay of a fictional ethical emergency scenario that is contextually situated in the team's specific workplace and applications. This paper outlines the DEED toolbox and describes three studies carried out with two different data science teams that iteratively shaped its design. Our findings show that practitioners can apply lessons learnt from the roleplay to real-life situations, and how the DEED opened up conversations around ethics and values.

Data Ethics Emergency Drill: A Toolbox for Discussing Responsible AI for Industry Teams

TL;DR

The findings show that practitioners can apply lessons learnt from the roleplay to real-life situations, and how the DEED opened up conversations around ethics and values.

Abstract

Researchers urge technology practitioners such as data scientists to consider the impacts and ethical implications of algorithmic decisions. However, unlike programming, statistics, and data management, discussion of ethical implications is rarely included in standard data science training. To begin to address this gap, we designed and tested a toolbox called the data ethics emergency drill (DEED) to help data science teams discuss and reflect on the ethical implications of their work. The DEED is a roleplay of a fictional ethical emergency scenario that is contextually situated in the team's specific workplace and applications. This paper outlines the DEED toolbox and describes three studies carried out with two different data science teams that iteratively shaped its design. Our findings show that practitioners can apply lessons learnt from the roleplay to real-life situations, and how the DEED opened up conversations around ethics and values.
Paper Structure (52 sections, 5 figures, 1 table)

This paper contains 52 sections, 5 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. Current Responsible AI and Its Shortcomings
  4. Designing for Reflection with Ethical Roleplay
  5. Research Context
  6. Author Roles
  7. Context of Participating Industry Teams
  8. Method
  9. What is a Data Ethics Emergency
  10. Research through Design
  11. Designing the DEED with practitioners
  12. Crafting the scenario
  13. Executing the drill session
  14. Evaluation - Data Collection and Analysis
  15. Post-Session Evaluation
  16. ...and 37 more sections

Figures (5)

  • Figure 1: Overview of Drill Studies - The cut off circles for scenario crafting represent that these activities were incomplete before the third drill. The dashed arrows pointing from the follow up interviews to the drills represent that participants from all drill iterations were included in the follow up.
  • Figure 2: Screenshot of example of values activity on the Miro board
  • Figure 4: Screenshot of example of activity for identifying communication channels on the Miro board
  • Figure 5: Screenshot of example activity for structuring the drill on the Miro board
  • Figure 6: Average Ratings of each study for perceived realism of the scenario, ability to provoke new thoughts and to elicit a useful discussion. The bars show the mean ratings and the black lines mark ranges of lowest rating and highest rating.