Skip to main content

Introduction

Instruction and application
Complete

Have you thought about the journey data takes from creation through transformation to the insights that power ML? That journey is data lineage—a core part of ethical data management.

Weak lineage, retention and metadata practices can introduce bias,errors andsecurity risk into models and undermine fairness and compliance. This unit shows how to evaluate policies and preprocessing so training data stays accountable.

Intro illustration

Why does this unit matter?​

Data is the foundation of ML and AI. Responsible management supports privacy, fairness and sustainability as systems grow more complex. You will learn how lineage, retention, metadata and preprocessing choices affect model reliability and organisational accountability.

Learning objectives​

By the end of this unit, you will be able to:

  • Analyse data lineage, retention and metadata practices to spot sources of bias and error.
  • Evaluate security policies and procedures that support fairness, transparency and accountability.
  • Design security protocols that protect sensitive information while enabling fair training.
  • Evaluate collection and preprocessing methods to minimise bias in datasets.

Prerequisite​

Complete Module 3 Unit 1: Ethical considerations in AI and ML before this unit.

Action item: Pause and think​

Before you continue, reflect on your own experience with data collection and management. Note examples where governance helped—or where gaps caused problems.

Reflection
Have you encountered biased or incomplete data in a project? How did it affect outcomes?
Your reflection here...
What steps do you take (or could you take) to handle data responsibly?
Your reflection here...
How can responsible data governance improve fairness and sustainability of AI models?
Your reflection here...