Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Modeling Fairness in Recruitment AI via Information Flow

Mattias Brännström, Themis Dimitra Xanthopoulou, Lili Jiang

TL;DR

The paper addresses fairness in AI-assisted recruitment by bridging technical and socio-ethical analyses through Information Flow Modeling (IFM), a framework that represents decision processes as networks of information sites and transformations. It applies IFM to a real-world recruitment pipeline combining AI-based matching with human decision making, constructing a multi-level information-flow model and tracing how biases can propagate to candidate outcomes. The study identifies specific bias sources, outlines downstream impact paths, and reveals where internal mitigations may fail, advocating for socio-technical governance and transparency enhancements. The work demonstrates IFM’s utility for structured fairness risk analysis, education, and governance, and points to future extensions such as scenario modeling and alignment with regulatory frameworks like the EU AI Act.

Abstract

Avoiding bias and understanding the real-world consequences of AI-supported decision-making are critical to address fairness and assign accountability. Existing approaches often focus either on technical aspects, such as datasets and models, or on high-level socio-ethical considerations - rarely capturing how these elements interact in practice. In this paper, we apply an information flow-based modeling framework to a real-world recruitment process that integrates automated candidate matching with human decision-making. Through semi-structured stakeholder interviews and iterative modeling, we construct a multi-level representation of the recruitment pipeline, capturing how information is transformed, filtered, and interpreted across both algorithmic and human components. We identify where biases may emerge, how they can propagate through the system, and what downstream impacts they may have on candidates. This case study illustrates how information flow modeling can support structured analysis of fairness risks, providing transparency across complex socio-technical systems.

Modeling Fairness in Recruitment AI via Information Flow

TL;DR

The paper addresses fairness in AI-assisted recruitment by bridging technical and socio-ethical analyses through Information Flow Modeling (IFM), a framework that represents decision processes as networks of information sites and transformations. It applies IFM to a real-world recruitment pipeline combining AI-based matching with human decision making, constructing a multi-level information-flow model and tracing how biases can propagate to candidate outcomes. The study identifies specific bias sources, outlines downstream impact paths, and reveals where internal mitigations may fail, advocating for socio-technical governance and transparency enhancements. The work demonstrates IFM’s utility for structured fairness risk analysis, education, and governance, and points to future extensions such as scenario modeling and alignment with regulatory frameworks like the EU AI Act.

Abstract

Avoiding bias and understanding the real-world consequences of AI-supported decision-making are critical to address fairness and assign accountability. Existing approaches often focus either on technical aspects, such as datasets and models, or on high-level socio-ethical considerations - rarely capturing how these elements interact in practice. In this paper, we apply an information flow-based modeling framework to a real-world recruitment process that integrates automated candidate matching with human decision-making. Through semi-structured stakeholder interviews and iterative modeling, we construct a multi-level representation of the recruitment pipeline, capturing how information is transformed, filtered, and interpreted across both algorithmic and human components. We identify where biases may emerge, how they can propagate through the system, and what downstream impacts they may have on candidates. This case study illustrates how information flow modeling can support structured analysis of fairness risks, providing transparency across complex socio-technical systems.

Paper Structure

This paper contains 25 sections, 1 equation, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Related work and IFM
  4. Situatedness and Participatory Approaches
  5. The Framework
  6. An Overview of the IFM Model
  7. Formal Definition of the IFM Model
  8. Alternatives
  9. Bias and Impact
  10. Method
  11. Results
  12. A summary of the recruitment process:
  13. Construction of the Information Flow
  14. Analysis of Biases and Mitigation
  15. Abstraction and Interpretation Biases
  16. ...and 10 more sections

Figures (2)

  • Figure 1: Overview of the recruitment process information flow corresponding to Table \ref{['tab:transitions']}. The included dashed green box displays the client process in where the channels a-g) is embedded. The channel AI Match is further detailed in Figure \ref{['fig:ai_match']}.
  • Figure 2: More detailed overview of AI Matching, corresponding to Table \ref{['tab:ai_match_breakdown']}.