Skip to main content

Introduction

Instruction and application
Complete

How much energy powers the world’s data centres? By 2030, some forecasts suggest data centres could use on the order of 900+ TWh per year—roughly the scale of national electricity systems.

AI and ML create value, but also environmental and social footprint: training cost, inference load, biased outcomes and governance gaps. This unit connects technical choices tosustainability and ESG expectations.

Intro illustration

Why does this unit matter?​

Large-model training can be carbon-intensive; organisations increasingly report AI alongside ESG goals. Professionals who can trade off accuracy, latency, cost and environmental impact will design systems that survive scrutiny from finance, risk and the public.

Learning objectives​

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

  • Analyse sustainable data products and how they relate to ESG responsibilities.
  • Design ML/AI approaches that include sustainability metrics and reduce negative environmental impact across the lifecycle.
  • Build frameworks for continuous sustainability assessment aligned with evolving standards.
  • Apply training and deployment strategies that balance performance with environmental considerations.

Prerequisites​

  • Async Unit 1: Ethical considerations in AI and machine learning
  • Async Unit 2: Data management for ethical and accountable ML

Action item: Pause and think​

Keep these questions in mind as you work through the lessons—they anchor the reflections below.

Reflection
What role does AI play in your organisation or industry? Are sustainability and ethics actively discussed?
Your reflection here...
How can organisations balance computational demand with environmental responsibility?
Your reflection here...
If you were asked to improve sustainability of an AI system tomorrow, where would you start?
Your reflection here...