The CPA Australia AI Hub

As a thinking and drafting partner

Example uses:

• Brainstorming approaches to client communication, training materials, or internal policies.
• Producing first drafts of emails, reports, cover letters, or engagement communications that you then edit for accuracy, tone and context.
• Translating technical accounting language into plain language explanations for clients or non-technical stakeholders.

Important to note: Use generative AI for first drafts and idea generation, not final outputs. You remain responsible for reviewing, editing and ensuring that any communication is accurate, appropriate and aligned with professional standards.

Supporting research and learning

Example uses:

• Clarifying unfamiliar concepts or standards before consulting authoritative sources.
• Suggesting search keywords, synonyms and approaches for more targeted research.
• Creating practice questions or scenarios for staff training on new standards, ethics or technology.

Important to note: Do not treat generative AI as an authoritative source on accounting standards, laws, or regulatory requirements. Always confirm information with primary sources.

Editing and proofreading

• Ask AI to highlight unclear wording, inconsistent terminology, or tone issues in your draft, rather than to rewrite it completely.
• Request explanations for suggested edits so you can improve your own writing skills.

APES 110 provides the ethical framework that guides the use of AI and other emerging technologies in professional practice. Whether using generative AI to analyse information, draft content, or support decision-making, members must continue to uphold the five fundamental principles of integrity, objectivity, professional competence and due care, confidentiality, and professional behaviour.

The technology-related revisions to APES 110 make it clear that responsibility for professional judgements and outputs remains with the accountant, not the technology. Members are expected to maintain appropriate technology-related skills, critically assess AI-generated information, apply professional judgement, and protect confidential information when using AI tools and systems.

The Code requires members to apply a conceptual framework to identify, evaluate and address threats to compliance with the fundamental principles.

When considering AI use, members should:

1. Identify potential threats (e.g. self-interest, familiarity, intimidation, advocacy) created or increased by using AI.
2. Evaluate the significance of those threats, using the “reasonable and informed third party” test.
3. Address threats by eliminating the activity, declining or ending the engagement, or applying safeguards.

Where generative AI has been used in a way that materially influences analysis, wording or conclusions, members should be prepared to explain how it was used, what checks were performed, and how compliance with APES 110 was maintained.

The effective use of AI requires more than individual judgement. It should be supported by appropriate governance arrangements, risk management practices, and systems of quality management that help ensure AI technologies are used responsibly, securely, and in accordance with professional obligations.

Before adopting or using an AI tool, organisations should consider factors such as privacy, security, data management practices, contractual arrangements, and the suitability of the technology for its intended purpose. The level of oversight should reflect the nature of the activity being performed, the sensitivity of the information involved, and the potential consequences of inaccurate or inappropriate outputs. While AI can improve efficiency and support decision-making, responsibility for professional judgements and final outputs remains with the accountant. AI-generated content, analysis, and recommendations should be subject to appropriate review and validation processes.

Organisations should also ensure that individuals using AI have the necessary knowledge and skills to understand the technology's capabilities and limitations, critically evaluate outputs, and apply professional judgement. Regular monitoring and assessment of AI tools can help identify risks relating to accuracy, bias, confidentiality, security, and legal or regulatory compliance.

Ultimately, AI should form part of an organisation's broader governance and quality management framework, supporting trustworthy outcomes while maintaining compliance with professional standards, ethical requirements, and stakeholder expectations.