By Clark Frogley, Head of Financial Crime Solutions at Quantexa
Online banking remains a prime target for sophisticated fraudsters, whose scams are as varied and adaptive as they are destructive. Banks find themselves in a relentless struggle, contending with innovative fraud tactics, major financial losses, and the erosion of customer trust.
The problem has just reached its peak as customer complaints in online banking have hit a 10-year high because of fraud, according to the UK’s Financial Ombudsman Service.
But with the right technology, banks can get ahead of this rise in fraudulent activity. Artificial intelligence can be used by banks to help combat money laundering and fraud at higher speed and with greater accuracy. But to achieve this, understanding the context of their data is key.
Building a strong data foundation
AI is a welcome addition to banks anti-fraud toolkits. But it’s only helpful if banks are able to drive the accuracy and reliability needed from AI to make informed and trusted decisions.
The big challenge is in unifying disparate and siloed data at scale to better understand customers and counterparties. By applying Entity Resolution to this data challenge, banks can bring together structured and unstructured data to build a 360-degree views of entities (individuals, organizations, and locations) and surface a deeper understanding of connections, relationships, and patterns in this data.
Entity Resolution enables banks to analyze the relationships between data, making it easier to spot anomalies in customer behavior which could signify fraudulent activity. With this technology, banks can convert vast quantities of low-quality data into meaningful, accurate descriptors of each entity that help to mitigate risk, optimize operations, improve the customer experience and accelerate revenue growth.
Using data to build customer trust in banking
To fend off fraud and rebuild declining customer trust, its vital banks have the technology in place that knows when to raise alerts for human involvement. Using contextual monitoring, banks are able to see a wider and more enriched view of the customer associated with any one transaction. From source of funds, right through to geo-location, this monitoring can raise the alarm based on contextual risk should a human need to take stock and control. These alerts are largely driven by both contextual aggravating and mitigating risk factors associated with the data. That means you don’t have to suffer large amounts of false positives to find the actual risk.
Banks, like HSBC are doing just this to improve the accuracy of fraud detection, using AI to ingest massive amounts of data and build networks using broader contextual information to zero in on real risk.
It’s crucial for banks to invest in new technology and tools in the ongoing race against fraudulent behavior allowing them to go from “guess-work” recommendations to informed and confident decisions derived from trusted data. In turn, offering them a steppingstone towards stronger fraud detection in banking.
