Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Insights for an AI Whistleblower Office from 30 Case Studies

Ethan Beri, Mauricio Baker

TL;DR

This paper takes an empirical approach, assembling a dataset of 30 case studies of whistleblowers, suggesting that whistleblower programmes will be more effective if they financially reward whistleblowers, provide protections for whistleblowers, enable whistleblowers to report anonymously, and provide advice to potential whistleblowers.

Abstract

Whistleblower programmes are a promising tool for uncovering noncompliance with AI regulations. This paper aims to help policymakers design an AI whistleblower programme by giving them an understanding of whistleblowers' motivations, and of the overall whistleblowing process. We take an empirical approach, assembling a dataset of 30 case studies of whistleblowers. This dataset includes dozens of features of each case, which range from 1978 to 2020 and span 15 industries. Our findings suggest that whistleblower programmes will be more effective if they financially reward whistleblowers, provide protections for whistleblowers, enable whistleblowers to report anonymously, are adequately staffed and funded, and provide advice to potential whistleblowers. We provide ten concrete policy recommendations for an AI whistleblower programme at the end of this paper.

Insights for an AI Whistleblower Office from 30 Case Studies

TL;DR

This paper takes an empirical approach, assembling a dataset of 30 case studies of whistleblowers, suggesting that whistleblower programmes will be more effective if they financially reward whistleblowers, provide protections for whistleblowers, enable whistleblowers to report anonymously, and provide advice to potential whistleblowers.

Abstract

Whistleblower programmes are a promising tool for uncovering noncompliance with AI regulations. This paper aims to help policymakers design an AI whistleblower programme by giving them an understanding of whistleblowers' motivations, and of the overall whistleblowing process. We take an empirical approach, assembling a dataset of 30 case studies of whistleblowers. This dataset includes dozens of features of each case, which range from 1978 to 2020 and span 15 industries. Our findings suggest that whistleblower programmes will be more effective if they financially reward whistleblowers, provide protections for whistleblowers, enable whistleblowers to report anonymously, are adequately staffed and funded, and provide advice to potential whistleblowers. We provide ten concrete policy recommendations for an AI whistleblower programme at the end of this paper.
Paper Structure (24 sections, 9 figures, 2 tables)

This paper contains 24 sections, 9 figures, 2 tables.

Table of Contents

  1. Executive summary
  2. Background
  3. Methodology and dataset
  4. Limitations
  5. Findings
  6. Demographics
  7. Organisations
  8. Wrongdoing
  9. Whistleblowing process
  10. Anonymity
  11. Retaliation
  12. Motivation
  13. Discussion
  14. Whistleblowers almost always appeared to be (at least partly) morally motivated ($\ge$87% of cases)
  15. Whistleblowers were almost always employees of the organisation which performed the wrongdoing ($\ge$90% of cases)
  16. ...and 9 more sections

Figures (9)

  • Figure 1: Prevalence of motivating (left) and demotivating (right) factors. Note we took a conservative, careful approach to ascribing motivation: each case used a range of sources, and we preferred a sparse field to an incorrect one. More on this in Methodology and dataset.
  • Figure 2: Whistleblower ages at the time of making their first report, bracketed (to include estimated ages, in cases where exact ages were unavailable).
  • Figure 3: Whistleblower education levels.
  • Figure 4: Wrongdoer organisations' industries.
  • Figure 5: Sizes of wrongdoer organisations.
  • ...and 4 more figures