MAS partners with banking industry to leverage AI and machine learning to fight financial crime

[SINGAPORE] The Monetary Authority of Singapore (MAS) announced on Monday (May 4) that it will leverage artificial intelligence and machine learning technology to fight financial crime.

To strengthen fraud detection capabilities, the central bank will work with banking industry partners, the Singapore Government Technology Agency and the Singapore Police Force.

MAS is conducting a proof of value (POV) exploring the use of AI and machine learning in preemptive fraud detection as part of a broader effort to apply these technologies to use cases across the industry.

MAS said the POV will not only lay the foundation for deeper industry collaboration, but also strengthen and complement financial institutions’ existing efforts to prevent and combat financial crime.

Additionally, POV added that it will use data from the five banks to build more robust and accurate AI and machine learning models that can identify high-risk transactions and accounts.

Such “rapid identification could enable timely assessment, intervention, and mitigation of customer losses due to fraud,” the agency noted.

To support POV, MAS provided its industry partners with a “secure data sharing environment governed by policies and protocols to protect customer information.”

Similarly, we have established a framework with industry stakeholders to ensure that the data shared is protected and used responsibly.

The data used in POV will remain confidential and will be protected by cryptographic technology, the central bank added.

Related items

For example, bank account numbers are hashed. This is an algorithmic process that replaces input data with a generated set of unique values. This ensures that only the contributing bank can identify the actual account number.

In addition, only authorized personnel can access data within a controlled setting that is constantly monitored through POV, MAS said.

All data used will be deleted at the end of the exercise.

After evaluating the effectiveness of POV, MAS said the scope and sophistication of the AI ​​and machine learning models used could be expanded by incorporating broader datasets and broader use cases.