Bank says new system can draft scam-detection rules while staff retain final approval
Commonwealth Bank has introduced an “agentic” artificial intelligence (AI) system intended to spot new scam and fraud patterns in payments data and to generate proposed detection rules designed to stop them.
The lender said the new AI capability sits within its broader fraud protection set-up, which it said monitors more than 80 million signals a day across transactions, card and online payments, and activity in digital banking channels.
The bank described the deployment as part of its annual $1 billion spending commitment aimed at protecting customers from fraud, scams, cyber threats and financial crime.
“When suspicious patterns are identified, the system quickly assesses their severity, analyses context, and proposes new detection rules to help intercept them,” said James Roberts (pictured right), executive general manager of fraud and scams at Commonwealth Bank.
“The new agent goes beyond traditional AI by not only rapidly identifying new threats but also determining how it can seek to disrupt them. The agent operates around the clock, continuously monitoring activity and adapting to emerging threats.”
Roberts said proposed rules are not applied automatically. He said they are checked and signed off by the bank’s fraud analytics team before being put into production, under a “human-in-the-loop” process.
The move comes as the bank investigates a large suspected mortgage fraud scheme that could involve as much as $1 billion in home loans. The alleged activity is understood to centre on falsified borrower information and other irregularities used to obtain approvals across a high number of mortgages. The bank is reviewing affected loans to assess the scale of any losses, identify accountability and determine possible legal or regulatory consequences.
Commonwealth Bank said it processes more than 20 million payments a day on average and sends more than 40,000 warning alerts to customers, on average, through the CommBank app. It claims that its detection technology helped cut fraud losses by more than 20% in the first half of the 2026 financial year compared with the first half of the 2025 financial year.
“The technology allows us to identify unusual events in highly complex patterns of activity at far greater speed and scale, helping us detect emerging threats sooner and update our controls faster,” Roberts said.
According to the bank, its data science and engineering teams built the system in three months, and that ongoing testing has shown strong detection results.
Want to be regularly updated with mortgage news and features? Get exclusive interviews, breaking news, and industry events in your inbox – subscribe to our FREE daily newsletter. You can also follow us on Facebook, X (formerly Twitter), and LinkedIn.
