Tech giant Google launched an anti-money laundering tool backed by AI on Wednesday for financial institutions to detect nefarious activities.

The tool uses machine learning (ML) to help financial institutions (FIs) identify risky transactions efficiently while driving down operational costs and improving consumer experience by using first-party data by the FI, according to Google’s release.

More than 95% of system-generated money laundering alerts are false positives, requiring manual review and leading to billions of dollars wasted on investigations each year, Anna Knizhnik, director of product management for, Cloud AI and Financial Services at Google Cloud, told Bank Automation News, quoting Thomson Reuters Regulatory Intelligence stats.

Anti-money laundering AI “minimizes wasted investigator time by reducing alert volumes by more than 60% and providing explainable outputs that speed up individual investigations,” Knizhnik said. “This reduction is achieved by lowering false positives while simultaneously accounting for additional financial crime risk.”

AML AI identifies high-risk transactions and has “outperformed incumbent systems across Europe, Latin America and Asia Pacific,” Knizhnik said.

The technology analyzes bank transactional patterns, network behavior and know-your-customer (KYC) data points to identify risk and is adaptive based on the underlying data.

AML AI also considers the movement of funds between geographies and jurisdictions, Knizhnik said. It learns from previously suspicious activities identified by the bank, whose policies dictate their approach to the risks they seek to mitigate.

All risk scores generated by AML AI are provided with ”explainability,” which defines the key risk indicators used to arrive at the score, Knizhnik added.

AML AI in action

HSBC, Bradesco and Lunar are among the FIs finding significant value in leveraging AI to track money laundering attempts, according to the Google release.

“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 of financial crime risk and compliance at HSBC said in the release. “We have been able to improve the precision of our financial crime detection and reduce alert volumes meaning less investigation time is spent chasing false leads.”

The London-based bank, which monitors more than 1.2 billion transactions monthly for financial crime, reported that AML AI helped detect “nearly two to four times more confirmed suspicious activity, eliminate over 60% of false positives” and “reduce over 60% of customer friction,” an HSBC spokesperson told BAN.