99% of business will use AI in financial reporting in 3 years: KPMG
Business leaders expect generative AI to revolutionise financial reporting, while auditors are being called upon to lead the transformation.
Within three years, nearly every Australian company will use AI in financial reporting while business leaders expect their auditors to be at the forefront of the transformation, according to a survey from KPMG International.
Already, nearly three-quarters of Australian businesses are using or piloting AI in financial reporting and the technology is claiming a bigger share of IT budgets each year.
The study surveyed global financial reporting executives and board members at 1,800 companies with revenues between $250 million and $1 billion.
Australia sized up positively for company AI adoption, placing third behind Canada and the UK.
Generative AI is still a fringe phenomenon, being used by only 9 per cent of Australian companies. That said, it is expected to overtake traditional AI and will be prioritised more than any other technology in financial reporting within three years.
“With the advent of gen AI, the entire ecosystem will evolve – requiring staff training and support, data management strategies to ensure compatibility and interoperability, investment in advanced analytic tools, and strong risk management processes to use gen AI responsibly and without bias,” said Elenie Carey, chief technology officer of audit and assurance at KPMG Australia.
“A system of continuous monitoring and improvement will be needed as the technology develops and the journey gathers pace.”
The main benefits of traditional AI flagged by Australian companies concerned anomaly detection and pattern recognition; robotic process automation; machine learning; and deep learning.
Gen AI is expected to mostly perform the same tasks as traditional AI, just more effectively and efficiently. Its deep learning capabilities mean it is better able to detect patterns and relationships between data points and can add a greater degree of actionability by recommending business actions.
“Technology skills are now essential for all auditors, not solely those specialising in IT,” wrote Bryant Ramdoo, partner and national audit and assurance innovation leader at KPMG Canada.
Among auditors, AI is quickly becoming less of a nice-to-have and more of a basic essential. More than three-quarters of business leaders globally said that AI is moderately to very important in the work of their external auditors.
Matt Campbell, chief technology officer in audit at KPMG UK, wrote that auditors are uniquely positioned to take a leading role in the transformation.
“Businesses are looking to their auditors to lead the AI transformation due to their deep understanding of financial reporting processes and ability to identify areas where AI can add the most value,” he wrote.
Boards of directors have even higher expectations for their auditors, with nearly two-thirds claiming it should be prioritised in identifying risks and anomalies, and 60 per cent preferring auditors use it in risk mitigation and internal controls.
“Just as AI has compelling applications in the financial reporting process, so too it brings powerful new capabilities to auditors,” said the report. “Businesses expect their audit companies to be in the vanguard of embracing and using it.”
Manuel Cortes, audit innovation partner at KPMG Spain, reported that the benefits of AI in auditing have begun to show.
“Our first experiences tell us that when auditors apply AI in their analysis of a company’s general ledger, they speed up their work, begin to obtain predictive analysis and more robust conclusions, and can propose improvements in the client’s internal processes to gain efficiency and minimise the risk of errors,” he said.
“AI can be a triple win for companies, auditors, and information users. It boosts quality, efficiency, and facilitates business decision making.”
More than half of the Australian respondents said their companies were devoting 10 to 20 per cent of their IT budgets to AI, while nearly one-third expect their AI spending to double within 12 months.
The main barriers to broader adoption of either form of AI include concerns around accuracy; cyber security; data organisation; transparency; data privacy; copyright and IP; data sovereignty; bias, and hallucinations.
Many of these concerns are more pronounced when it comes to generative AI, notably, when it comes to cyber security and data privacy.
