AI Modes, Governance, and Data Integrity

This unit explores operational modes of AI (AI‑First, Fusion, Human‑First, Marginal), strategies for transitioning among them, and governance practices that preserve human agency in financial services. Students learn humanistic reinforcement approaches, algorithmic recalibration methods for safe model adaptation, and the principles of data as a strategic, regulated asset to prepare for deployment and ongoing oversight.

Learning Objectives

• Analyze the characteristics, benefits, and risks of AI‑First, Fusion, Human‑First, and Marginal modes and recommend transition strategies for financial services use cases.
• Apply humanistic reinforcement and human‑in‑the‑loop controls to design AI workflows that maintain explainability, accountability, and human agency in decision making.
• Demonstrate algorithmic recalibration processes—monitoring, bias detection, retraining triggers, and version control—to safely adapt models in production environments.
• Evaluate data as an organizational asset by assessing data quality, provenance, privacy, and governance, and propose safeguards and value‑capture strategies for AI implementations in finance.