As consumers embrace ecommerce, digital banking, and online payment applications, the risk of fraud and other financial crimes has increased dramatically. Every new portal and mobile app expand the attack surface and give hackers new opportunities to exploit vulnerabilities. The stakes for financial organizations are growing as well. In 2021, U.S. fraud losses amounted to $5.9 billion, a 436% increase over 2017 levels, according to McKinsey. And for every dollar lost to fraud, banks spend over $4 on recovery fees, legal fees, and other expenses.
Thwarting financial crime is never easy, but by adopting the right cloud infrastructure and strategically deploying artificial intelligence (AI) technologies, financial institutions can get ahead of bad actors, gaining insight into their tactics, discovering their activity sooner, and preventing attacks before they lead to a loss.
Challenges for fraud risk management
Fraud is a big and a worthwhile business for today’s online criminals, who troll the internet and insert data-stealing malware into vulnerable sites and mobile apps. They sell stolen data on the dark web, where they form alliances to trade tactics and technologies, such as AI algorithms that can crack even the most complex passwords in seconds. They purchase thousands of illicit debit and credit card numbers and combine them with other hacked information—including Social Security numbers of children, who have no financial history—to create false identities and open multiple fraudulent accounts.
To counter these measures, financial institutions are taking increasingly aggressive steps to identify suspicious behaviors and transactions quickly.
“One main challenge faced by financial organizations is the complexity of correlated data needed to discover deviant behavior and patterns,” says Archana Trikha, Associate General Manager at Tata Consultancy Services (TCS). “Mature banks perform massive volumes of transactions, but less than one percent are fraudulent, so it’s basically finding the needle in the haystack. When a card or an ATM transaction is taking place, you need to identify suspicious activity and take a decisive action in milliseconds.”
Traditional tables and relational databases at the back end of financial systems and applications are often siloed. By missing important correlations, analysts or automated systems may not spot illegal activity soon enough to prevent a loss. When suspicious activity is flagged, it can hold up a transaction while a representative investigates the matter, frustrating customers.
Fighting back with AI
“Effective fraud prevention requires financial institutions to stay a step ahead of cybercriminals by incorporating AI technologies and multiple dimensions related to sequence of events to identify suspicious activity in real time,” Trikha says. One essential step is accessing the processing power of Microsoft Cloud, where organizations can integrate and analyze large data sets from multiple data sources, swiftly correlating them for possible connections to fraudsters or money-laundering shell companies.
For example, organizations can use cloud-based AI applications to analyze customer behavior, discerning near-real time deviations from a real customer’s profile, identifying anomalous behavior, or identifying collusions among entities. AI can also enhance existing software, such as alerting systems, which often generate many false positives. In addition to reducing alert fatigue, AI capabilities can provide reasons for its decisions, saving staff hours of time in compiling compliance reports.
“These are just some of the advanced technologies TCS deploys and customizes for financial clients,” Trikha says. Once big data and AI systems are set up in Microsoft Cloud, they continue to incorporate the latest information about threats, investigations, and interventions, helping financial institutions to stay one step ahead of criminals.
Learn how to master your cloud transformation journey with TCS and Microsoft Cloud.
