Google Cloud launches AI-powered AML tech

Google Cloud has developed an AI-based product designed to help global financial institutions more effectively detect money laundering.

According to the United Nations, although banks devote enormous resources to this problem, the amount laundered each year is estimated at 2% and 5% of global GDP, or $2 trillion per year.

Google claims that most older AML monitoring products are based on manually defined rules, resulting in a low rate of identifying suspicious activity. In fact, more than 95% of system-generated alerts are “false positives” during the first investigation phase, with 98% never resulting in a Suspicious Activity Report (SAR).

Google Cloud Responses provides machine learning-generated aggregate customer risk scores as an alternative to rule-based trading alerts. The risk score is based on banking data, including transaction patterns, network behavior, and KYC data to identify high-risk retail and commercial customer cases and groups.

The company claims that its AML AI technology not only increases risk detection, but also reduces operating costs and improves customer experience.

HSBC uses the system, reporting it has helped identify two to four times more suspicious activity, while reducing alert volume by more than 60%.

“Google Cloud’s AML AI has significantly improved HSBC’s AML detection capability. Google’s models are already demonstrating the tremendous potential of machine learning to transform anti-financial crime efforts in the industry at large,”

Jennifer Calvery, group head, financial crime risk and compliance, HSBC.