A KPMG partner paid a Rs 6.4-lakh fine for cheating in an internal AI test—by using AI

Andrew Yates, Chief Executive, KPMG Australia, admitted that managing AI use within organisations has become increasingly complex. “Like most organisations, we have been grappling with the role and use of AI as it relates to internal training and testing. It’s a very hard thing to get on top of given how quickly society has embraced it,” he said.

“Given the everyday use of these tools, some people breach our policy. We take it seriously when they do. We are also looking at ways to strengthen our approach in the current self-reporting regime,” he added.

The issue reflects a wider challenge across the accounting world. In December, the Association of Chartered Certified Accountants (ACCA), the UK’s largest accounting body, announced that it would move its exams to in-person settings, saying online safeguards were no longer enough to curb AI-assisted cheating. ACCA chief executive Helen Brand described the moment as a “tipping point”, noting that AI capabilities were advancing faster than systems designed to maintain exam integrity.

Ironically, firms such as KPMG and PricewaterhouseCoopers have been encouraging staff to use AI tools in their day-to-day roles to boost productivity and reduce costs. At KPMG, partners are even expected to be assessed on their ability to leverage AI during their 2026 performance reviews.

This is not the first time the firm has faced scrutiny over testing practices. In 2021, KPMG Australia was fined AUD 615,000 after more than 1,100 partners were found to have engaged in “improper answer-sharing” in professional skills and ethics assessments.

The company says it has since tightened oversight and will continue tracking how employees use AI, aiming to strike a balance between embracing new technology and maintaining professional standards.