Responsible AI in Practice

Human oversight and accountability for automated decisions.

Move beyond principle lists. Examine where responsibility sits, what meaningful human review requires and how organisations can make AI-assisted decisions explainable, challengeable and governable.

Learning outcomes

What you will learn to do.

Develop the legal judgement needed to question, govern and document the use of AI in consequential decisions.

01
Explain why organisational responsibility remains with the people and institutions that design, procure or use an AI system.
02
Identify risks including discrimination, privacy intrusion, opacity, unsafe reliance and barriers to challenge.
03
Distinguish meaningful human oversight from routine approval or rubber-stamping.
04
Map responsibility across providers, deployers, vendors, professionals, users and decision owners.
05
Recognise when review, escalation, documentation, contestability and intervention are required.
06
Apply a practical governance framework to an automated decision-making use case.

Course modules

Ten modules of guided study.

Each module connects legal and governance questions to decisions, documents and practical work.

Module 1

What Responsible AI Requires

From principles to decisions, controls and evidence.

Module 2

AI Systems, Roles and Responsibility

Providers, deployers, vendors, users and decision owners.

Module 3

Risk Classification and Impact Assessment

Context, affected people, foreseeable harm and proportional controls.

Module 4

Data, Bias and Discrimination

How data and system design can create or compound unequal outcomes.

Module 5

Transparency, Explanation and Contestability

Giving people useful information and a realistic route to challenge.

Module 6

The Human in the Loop

Meaningful oversight and accountability for automated decisions.

Includes the thought-leadership session: The Human in the Loop: Accountability in Automated Decision-Making.

Module 7

Privacy, Security and Sensitive Uses

Purpose, access, retention, security and higher-risk contexts.

Module 8

Procurement and Third-Party AI

Diligence, contractual allocation, assurance and supplier evidence.

Module 9

Monitoring, Incidents and Change

Drift, complaints, escalation, rollback and post-deployment review.

Module 10

Capstone: Responsible AI Governance Plan

A usable governance plan for a defined system and organisation.

Course capstone

Build a governance approach you can use.

Complete the pathway with a responsible AI governance plan for a defined system, covering roles, risk controls, oversight, challenge and post-deployment monitoring.