Skip to content
Français

AI Governance: Policies, Risk and Accountability

Set the rules for AI in an organisation, give every AI system an owner, and prove that the rules are followed. You govern Larkfield's AI portfolio, read the main laws and standards in plain words, and finish with a governance pack for the next year.

Read every lesson as a guest, or sign in to enroll and track your progress.

What you'll learn

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

  • Explain what AI governance is for, and set accountability so that every AI system has an owner and every decision has an approver.
  • Explain in plain words what the main laws and standards ask (the EU AI Act, the NIST AI Risk Management Framework, ISO/IEC 42001 and 42005, and the situation in Canada and other places), and when to ask a specialist.
  • Build an AI inventory, set risk tiers and run an impact assessment in proportion to the risk.
  • Write policies that staff and teams can follow: acceptable use, building and buying AI, and vendor checks.
  • Turn policy into controls and documentation, and collect evidence that an auditor would accept.
  • Run governance week to week: a board, fast paths for low risk, incidents, complaints and redress, and measures that show governance helps.

About this course

Larkfield’s platform team finds a script that nobody registered. Each week it sends customer messages to an AI service that Larkfield has no contract with. At the same time, the support team wants its AI workflow to refund small amounts with no person’s approval, and three tools that Larkfield buys have added AI features without asking. Larkfield sells to customers in Canada and in the European Union. Who decides what is allowed? Who is accountable when something goes wrong? And how does Larkfield show a customer, an auditor or a regulator that it follows its own rules?

In this course, you follow Farida, Larkfield’s first risk and compliance lead. You build the AI inventory, set risk tiers and write an impact assessment. You write policies that staff can follow without a lawyer, check vendors, and turn each rule into a control with evidence. You run a governance board, and you handle an incident and a complaint.

You read policies, tables and short reports, never code. The laws and standards are explained in plain words, each with its official source and the date it was checked. This course is not legal advice: each lesson on law says what to ask a lawyer. Every number has a label: recorded in an earlier course, an official source, a stated assumption, or a constructed example. At the end, you write Larkfield’s governance pack for its next year.

Before you start

None. You must be able to read a short policy, a table and a simple report. You do not write code, and you need no account and no money. You need a browser and a document tool (a word processor, a text editor or a spreadsheet) to fill in the course's templates. This course explains laws in plain words; it is not legal advice.

Who this is for

People who set or check the rules for AI in an organisation. Business and product leaders, risk, compliance and privacy staff, internal auditors, security and IT leads, and founders of small companies who do all of these jobs.

How we made this course

Experienced AI practitioners designed this course. AI helped us draft the content; people chose the teaching approach and the examples, and reviewed every lesson. The outputs that you see are the actual results of running the code. How we make our courses

Inside the course

Module 3Inventory, risk tiers and impact assessment

3 lessons

Inventory, risk tiers and impact assessment. List every AI system, including the AI inside tools you buy, set a risk tier for each one with the controls it needs, and write an impact assessment for a high-tier system.

  1. The AI inventory30 min
  2. Risk tiers and proportionate controls35 min
  3. Impact assessments35 min

Module 5Controls, documentation and evidence

3 lessons

Controls, documentation and evidence. Turn each policy line into a control with an owner and a proof, keep system cards and decision records current, and answer an auditor's requests from the evidence.

  1. From policy to control30 min
  2. Documentation30 min
  3. Evidence and audits35 min

Final module · choose at least oneCase studies

3 case studies

Three case studies on public records. Review a tribunal decision about a chatbot's answer, write the impact assessment that a public grading algorithm needed, and build an inventory from published algorithmic impact assessments. Complete at least one for the certificate; each one is a portfolio piece.

Complete at least one case study for the certificate. The others are optional. What each one covers

  1. A tribunal case about a chatbot's answer120 min
  2. Learn from a public algorithm failure120 min
  3. Review a published AI register120 min

After the case studiesFinal assignment

Use every module in one practice project. You collect evidence, and check it against the course outcomes.

  1. Final assignment180 min

Case studies

Realistic projects from start to finish, on real data, with every output explained. Each one makes a portfolio piece. Complete at least one for the certificate; the others are optional.

  1. A tribunal case about a chatbot's answer2 hr

    Read a public tribunal decision about a company held responsible for what its chatbot told a customer, and write the governance review: the policy, owner, review and record that would have prevented it.

  2. Learn from a public algorithm failure2 hr

    Read the official reports on a public-sector automated-decision failure, and write the impact assessment that should have been done before launch.

  3. Review a published AI register2 hr

    Use published algorithmic impact assessments under an open licence to build an inventory and risk tiers, and find what the published records leave out.

Reviews

Loading the reviews…

    Took this course? Sign in to review it. Learners can write a review after they complete 25% of the lessons.

    Share this course