Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Operationalizing Justice: Towards the Development of a Principle Based Design Framework for Human Services AI

Maria Y. Rodriguez, Seventy Hall, Pranav Sankhe, Melanie Sage, Winnie Chen, Atri Rudra, Kenny Joseph

TL;DR

The paper tackles how to embed justice into high-stakes human services AI by applying Value Source Analysis to administrative policy, using New York State as a case study. It employs a computational grounded theory pipeline—combining structural topic modeling, qualitative coding, and embedding-based validation—to surface 13 justice principles and 8 AI design guidelines. The findings illuminate how policy defines and constrains what justice means in practice, proposing a policy-informed framework for designing AI in child welfare and other human service contexts. The work advances a pragmatically bounded approach to ethical AI, emphasizing alignment with existing policy values and offering guardrails to mitigate harm in high-stakes decisions.

Abstract

Scholars investigating ethical AI, especially in high stakes settings like child welfare, have arguably been seeking ways to embed notions of justice into the design of these critical technologies. These efforts often operationalize justice at the upper and lower bounds of its continuum, defining it in terms of progressiveness or reform. Before characterizing the type of justice an AI tool should have baked in, we argue for a systematic discovery of how justice is executed by the recipient system: a method the Value Sensitive Design (VSD) framework terms Value Source analysis. The present work asks: how is justice operationalized within current child welfare administrative policy and what does it teach us about how to develop AI? We conduct a mixed-methods analysis of child welfare policy in the state of New York and find a range of functional definitions of justice (which we term principles). These principles reflect more nuanced understandings of justice across a spectrum of contexts: from established concepts like fairness and equity to less common foci like the proprietary rights of parents and children. Our work contributes to a deeper understanding of the interplay between AI and policy, highlighting the importance of operationalized values in adjudicating our development of ethical design requirements for high stakes decision settings.

Operationalizing Justice: Towards the Development of a Principle Based Design Framework for Human Services AI

TL;DR

The paper tackles how to embed justice into high-stakes human services AI by applying Value Source Analysis to administrative policy, using New York State as a case study. It employs a computational grounded theory pipeline—combining structural topic modeling, qualitative coding, and embedding-based validation—to surface 13 justice principles and 8 AI design guidelines. The findings illuminate how policy defines and constrains what justice means in practice, proposing a policy-informed framework for designing AI in child welfare and other human service contexts. The work advances a pragmatically bounded approach to ethical AI, emphasizing alignment with existing policy values and offering guardrails to mitigate harm in high-stakes decisions.

Abstract

Scholars investigating ethical AI, especially in high stakes settings like child welfare, have arguably been seeking ways to embed notions of justice into the design of these critical technologies. These efforts often operationalize justice at the upper and lower bounds of its continuum, defining it in terms of progressiveness or reform. Before characterizing the type of justice an AI tool should have baked in, we argue for a systematic discovery of how justice is executed by the recipient system: a method the Value Sensitive Design (VSD) framework terms Value Source analysis. The present work asks: how is justice operationalized within current child welfare administrative policy and what does it teach us about how to develop AI? We conduct a mixed-methods analysis of child welfare policy in the state of New York and find a range of functional definitions of justice (which we term principles). These principles reflect more nuanced understandings of justice across a spectrum of contexts: from established concepts like fairness and equity to less common foci like the proprietary rights of parents and children. Our work contributes to a deeper understanding of the interplay between AI and policy, highlighting the importance of operationalized values in adjudicating our development of ethical design requirements for high stakes decision settings.

Paper Structure

This paper contains 24 sections, 2 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Literature Review
  3. A brief review of child welfare AI applications
  4. Child Welfare Law and Values
  5. Value Sensitive Design and Policy Analysis
  6. Data and Methods
  7. Collecting Child Welfare Policy in the United States
  8. Initial Coding of New York State Policy
  9. Mixed Methods Analysis to Surface Principles of Justice
  10. Positionality Statement
  11. Results
  12. Defining Justice
  13. Communicating Justly
  14. Intangible Justice
  15. Acting Justly
  16. ...and 9 more sections

Figures (2)

  • Figure 1: Descriptives of the dataset on a per-state basis. Presented are the total number of words across all state statutes (left subplot) and the total number of administrative statutes (right).
  • Figure 2: A word cloud representing one topic from our initial topic modeling of all state child welfare policy