AI Governance, Risk & Compliance for Finance and Business
When your team asks, “Can we use AI for this?”
You need an answer that considers the work, the information and the consequences. A tool approval or a general policy rarely answers every question.
This course gives finance and business professionals a practical method for assessing AI use. Work through realistic proposals, challenge vendor claims, review failed controls and prepare a recommendation a CFO, partner or manager can act on.
You finish with a governance decision pack you can adapt to your workplace, including a use register, risk assessment, approved-use matrix, policy draft and management update.
What you will be able to do
- ✓Identify AI uses, owners, data and actions across a team or client engagement.
- ✓Decide which proposals need closer assessment, a limited pilot, a hold or redesign.
- ✓Ask suppliers for evidence that addresses the actual service and information involved.
- ✓Specify meaningful human review and check whether access and approval controls work.
- ✓Write practical operating rules, respond to incidents and reassess changes.
- ✓Report benefits, errors and unresolved risks without overstating what the evidence shows.
Practise the decisions that come up at work
Reporting and review
Assess a management-reporting assistant, check review coverage and decide whether payroll access or direct emailing is within scope.
Invoice processing
Review a supplier evidence pack, identify critical control failures and decide whether an invoice assistant is ready for real data.
Client and customer impact
Examine engagement-file separation, an accidental disclosure and an account-suspension proposal that needs redesign.
All records are fictional and supplied. Each exercise tells you what to do, what to save and how to compare your work with a worked answer.
Who this course is for
Finance managers, financial controllers, CFOs, business managers, partners, practice managers and risk or compliance professionals who help decide how AI may be used.
The examples cover both in-house teams and accounting or advisory firms. You can take the course independently, whether you are assessing your first team proposal or improving an existing approval process.
You should understand everyday business processes and professional review. No coding, legal qualification or previous Academy course is required.
Your learning path
▸Module 1: Know what is being used and who is accountable35 minutes · An AI use register and a clear allocation of responsibilities.
- ✓1.1 Describe the work before discussing the tool
- ✓1.2 Identify which obligations apply
- ✓1.3 Give each decision a responsible person
▸Module 2: Assess the use, the data and the possible harm40 minutes · A risk assessment that leads to a practical decision.
- ✓2.1 Separate a useful draft from an action that affects people
- ✓2.2 Write a risk statement you can act on
- ✓2.3 Make human review work in practice
▸Module 3: Approve the tool, workspace and data together45 minutes · A vendor evidence review and an approved-use matrix.
- ✓3.1 Check what information will enter and leave the service
- ✓3.2 Turn vendor claims into evidence requests
- ✓3.3 Publish an approved-use matrix people can read
▸Module 4: Test controls before you approve a pilot45 minutes · A test review, a realistic time calculation and an approval recommendation.
- ✓4.1 Decide what a control must demonstrate
- ✓4.2 Read the results without hiding a serious failure
- ✓4.3 Balance benefit, effort and the approval conditions
▸Module 5: Write rules people can apply at work40 minutes · A short operational policy, rollout plan and change process.
- ✓5.1 Draft an operational AI policy
- ✓5.2 Check that staff can use the rules
- ✓5.3 Reassess changes before expanding use
▸Module 6: Respond when AI use goes wrong35 minutes · An incident log, a factual update and conditions for restart.
- ✓6.1 Contain the problem and establish the facts
- ✓6.2 Prepare a factual update and route the notification decision
- ✓6.3 Set evidence-based conditions for restart
▸Module 7: Report clearly and complete your decision pack60 minutes · A management update, final practical pack and knowledge check.
- ✓7.1 Give management a clear view of performance and exposure
- ✓7.2 Assemble your final governance decision pack
- ✓7.3 Complete the final knowledge check
Approximately 5 hours including reading, guided activities, review and the final assessment. Study times are estimates.
A working pack to take back to your organisation
- ✓An AI use register and an obligations map.
- ✓A risk and control assessment, vendor questions and an approved-use matrix.
- ✓A test review and documented approval recommendation.
- ✓An operational policy draft, staff scenarios and a change record.
- ✓An incident response log, management update and final evidence index.
The downloadable workbook is editable plain text that you can use in a text editor or copy into your preferred document application. Source records and worked answers are included.
You will practise how to adapt the documents to the decision. They support professional judgement and do not replace your organisation's approval process or specialist advice.
What you need
A computer, a document or text editor and a calculator. No paid AI subscription is required. You review supplied records and prepare governance decisions, so you do not need access to a live client system.
The approach applies when assessing ChatGPT, Claude, Microsoft services and other AI products. The course does not depend on one provider's interface or a particular model version. It teaches you to verify the actual service, plan, settings and contract.
Use only the supplied fictional material for course activities. Do not upload client, employer, personal or confidential information for the exercises.
Complete the course and earn your badge
Build your practical governance decision pack and complete the final knowledge check. Answer all 10 questions to see your result. The pass mark is 80% (8 out of 10) , with explanations to help you review and try again.
On completing the course requirements through the Academy, you receive a certificate of completion and a badge for this course . The award recognises your learning; it is not a professional licence or certification of your organisation's compliance.
Created with a finance professional’s perspective
Angela Shi
Founder & CEO, Empathetic AI · Former fintech CFO · Finance educator
Angela brings more than 15 years of experience across financial services and FinTech, alongside her work developing AI applications and teaching business and finance professionals.
The course focuses on the questions finance leaders face in practice: what can be approved, which evidence is missing, who remains accountable and what should happen when a control fails.
Developed by Empathetic AI, an Australian AI company working at the intersection of finance expertise, practical implementation and education.
Questions before you enrol
▸Is this a legal or technical qualification?
It is practical professional education in AI governance. You learn to prepare assessments, test reviews and decisions. It does not qualify you to give legal advice or certify an organisation against a management-system standard.
▸Do I need to take the other Academy courses first?
No. This course is independent and includes its own explanations, practice files, assessment, certificate and badge.
▸Will I need ChatGPT Enterprise, Claude or Microsoft Copilot?
No paid AI account is needed to complete the activities. You assess the supplied fictional evidence. The vendor-review method can be applied to the specific product and plan your organisation proposes.
▸Is the course relevant to Australian organisations?
Yes. It uses an Australian context and links to primary government, privacy, accounting and tax-practitioner sources. You learn to check which requirements apply rather than treating every organisation as subject to the same rules.
▸Can I use the workbook at work?
Yes. Adapt it to the actual use, authority, policies and obligations in your organisation. Obtain the required review before introducing a policy or approving an AI use.
▸How long should I allow?
Approximately 5 hours, including reading, activities, calculations, practical review and the final knowledge check. You can work through it at your own pace.
Prepare your next AI approval with a clear method.
Work through the evidence, make the decision and leave with a pack you can explain.