Bias and Valuation in AI

This unit examines the nature, sources, and consequences of bias in AI systems used in financial services, including paradoxes around the pursuit of ‘impartial’ data and the distinction between specific and systemic biases. Students will synthesize chapters 1–8 of the course references to evaluate how societal valuations of human life influence model outcomes and to prepare evidence-based ethical mitigation strategies for subsequent implementation work.

Learning Objectives

• Analyze the paradoxes of striving for impartial data and differentiate specific versus systemic biases as they arise in financial services AI scenarios
• Evaluate the dual nature of biases (individual and systemic) and assess their measurable impacts on model decisions and affected stakeholder groups using course case examples
• Apply ethical frameworks and mitigation strategies to critique how AI systems encode societal valuations of human life and propose prioritized interventions to reduce harm in financial decision-making
• Demonstrate the ability to synthesize and cite chapters 1–8 from the course references to support evidence-based assessments and recommendations for addressing bias in AI systems