CommBank Develops Agentic System for Building Defences Through AI Fraud Detection

Commonwealth Bank launches an advanced agentic AI system to monitor 80 million daily signals, reducing fraud losses by 20% through real-time scam pattern detection.

Commonwealth Bank recently announced that it was deploying an advanced agentic AI system that would help with AI fraud detection and scam patterns in both payments and transaction data. This also helps generate the rules needed to help intercept them. Investing in this system will form a part of CommBank’s $1 billion annual commitment to mitigate the threats of fraud, scams, cyber threats, and financial crime for customers. The bank’s advanced fraud protection systems consist of various AI capabilities that monitor more than 80 million signals each day, including transactions, card and online payments, along with interactions with digital banking channels.

Furthermore, elaborating on the agent’s capabilities of fraud detection using AI, Commonwealth Bank’s Executive General Manager of Fraud and Scams, James Roberts, noted that it bolsters the bank’s ability to respond rapidly to evolving threats by continuously monitoring activity around the clock and adapting to emerging threats accordingly. The bank’s fraud analytics team reviews and approves new detection rules before they are implemented in a process known as human-in-the-loop oversight. With the bank processing more than 20 million payments on average, its fraud detection technology has helped reduce fraud losses by over 20% in the first half of 2026 compared to the first half of 2025.

Moreover, the agent has also assisted in developing and updating a large part of the Commonwealth Bank’s card fraud rules used to identify potential fraud. Also, the bank’s in-house data science and engineering teams have developed the fraud detection system in three months, and the ongoing testing validates the efficacy of the same. According to CIO Bulletin, this initiative rightfully reinforces Commonwealth Bank’s position as among the top four banks globally for AI maturity. The bank’s technology helps it to identify unusual and complex patterns quickly for detecting emerging threats and take necessary corrective measures.