Algorithmic Bias

Algorithms do not become unfair by accident. They learn from the world we give them. In hiring, credit, healthcare, policing, education, and public services, automated systems increasingly shape human lives — and when those systems inherit distorted data, proxy variables, and weak oversight, they can reproduce injustice at scale.

In Algorithmic Bias, Flávio Napoleão explains how unfairness enters AI through data, design, law, and governance — and what institutions must do to prevent it, drawing on real cases from facial recognition, hiring algorithms, credit scoring, and criminal risk assessment.

Table of contents

1. The Mirror That Lies
2. How Machines Learn Our Worst Habits
3. The Hiring Machine
4. The Accuracy Paradox
5. When the System Says No
6. What the Machine Inherits
7. Accountability and Recourse
8. The Decision Not to Build
9. Design for Contestability
10. The Architects of Trust

More from this book

• Algorithmic Bias Meets Privacy Law: Where the EU AI Act and GDPR Overlap
• Algorithmic Bias Glossary: COMPAS, Proxy Bias, Contestability, and Other Terms You Need to Know
• The Architects of Trust: Who Should Be in the Room When AI Gets Approved?
• The Accuracy Paradox: Why a 'Statistically Fair' Algorithm Can Still Be Unjust
• How Machines Learn Our Worst Habits: The Hidden Bias in Classification Systems
• Design for Contestability: What Michigan's MiDAS Disaster Teaches About AI Accountability
• The Decision Not to Build: Lessons from the Netherlands' SyRI Welfare Scandal
• What the Machine Inherits: The SCHUFA Case and the Myth of Neutral Data
• Inside COMPAS: How ProPublica Exposed the Algorithm Sentencing America
• When the System Says No: A Lawyer's Own Encounter with Algorithmic Refusal
• The Amazon Hiring Algorithm That Learned to Discriminate Against Women
• The Robert Williams Case: What Facial Recognition Bias Really Costs