AI in Financial Reporting: New Frontiers of Risk Exposure

Artificial intelligence (AI) is rapidly reshaping our lives and changing the way businesses are delivering their services. However, despite the potential benefits, AI introduces a range of risks, particularly in the context of financial reporting. Whether through simple error or as part of a more complex fraud, AI can contribute to financial misstatements in ways that are not always immediately apparent.

Overreliance on AI and judgement risk

One of the most significant risks is overreliance on AI in areas that require professional judgement and estimation. Small and medium-sized entities are increasingly using AI to support technical accounting processes, including valuations, provisions, impairment, and other subjective areas that rely heavily on judgement.

AI systems are trained on historical data, which may contain biases or outdated assumptions. As a result, calculations can deviate significantly from economic reality. Skewed estimates in these areas can have a material impact on financial statements, underscoring the importance of human oversight and professional scepticism when reviewing AI-generated outputs.

AI in regulatory and compliance decisions

Beyond technical calculations, AI is also being used to guide regulatory and compliance decisions. Entities may rely on AI tools to interpret complex reporting requirements, draft disclosures, or determine accounting treatments.

However, these tools often provide generic or inaccurate guidance that does not account for jurisdictional nuances or specific organisational circumstances. Without proper oversight, such reliance can lead to incorrect accounting treatments and non-compliance with applicable standards. This is particularly the case where management lacks the expertise or time to critically evaluate AI recommendations.

Impact on internal controls

While AI can significantly streamline time-intensive processes, this efficiency should not come at the expense of strong internal controls. Increased automation can inadvertently weaken control frameworks if not properly designed and monitored.

For example, if AI systems are relied upon too heavily without appropriate checks, errors may go unnoticed and flow through downstream processes. It remains critical that fundamental principles such as segregation of duties, proper authorisation, and independent review are maintained across key business functions such as accounts payable and payroll. AI should be implemented in a way that supports control environments, rather than replacing essential human oversight and accountability.

Data security considerations

The use of AI also introduces data security considerations, particularly where confidential business or financial information is entered into external platforms. Popular AI tools such as ChatGPT may retain user inputs to improve their models, meaning sensitive company information could be exposed if not handled carefully.

As a result, businesses should implement appropriate safeguards and ensure employees are clearly briefed on the appropriate use of AI.

Fraud and manipulation risk

Perhaps most concerning is AI’s potential to facilitate fraud, thereby distorting financial reporting outcomes. Advances in generative AI have made it increasingly easy to produce convincing fake documentation including invoices, contracts and emails.

These documents can be used to support fictitious transactions, inflate revenue, conceal liabilities, or justify unsupported balances within the financial statements. Without effective audit work, the sophistication of these tools makes it difficult for both the board of directors and shareholders to distinguish between genuine and manipulated business outcomes. This elevates the importance of robust verification procedures, data analytics, and testing across multiple sources to identify anomalies and mitigate fraud risk.

Managing AI risk through governance

These threats may be mitigated through proactive strategy and strong governance processes. Our team supports organisations navigating these emerging risks, with a focus on maximising the benefits of AI while implementing robust processes and controls to mitigate risk and uphold the integrity of financial reporting.

If you would like to talk through how this may apply to your organisation, feel free to reach out to your Accru Adviser.

About the Author
Nicholas Apostolou, Melbourne
Nicholas is a Graduate Auditor at Accru Melbourne, working alongside clients in diverse industries to support audit and compliance engagements. He is committed to precision, ongoing professional development, and building strong financial foundations for the businesses he works with.
Want to join the team?