Deep Dive: Fighting Fraud at the Genetic Level |
Digital technology has revolutionized the way customers engage with their banks and technological advancements are granting individuals and businesses new ways to manage their finances and access more personalized services The increasingly digital nature of banking has also opened new avenues for fraudsters Customers are now performing high volumes of transactions that are processed automatically giving criminals seemingly limitless chances to commit fraud This is particularly true in the European market which recently experienced new developments that could leave it vulnerable Open banking in the UK and the updated Payment Services Directive PSD2 encourage greater competition in the financial services sector by requiring banks to share consenting customers information with third parties Companies seeking to comply with these new standards risk making themselves vulnerable to bad actors This potential for wrongdoing presents a twofold challenge for banks First FIs must work aggressively to detect fraudulent activities by investigating and anticipating suspicious behavior Second they must take steps that do not negatively affect their customers experiences AI ML and risk management tools will play significant roles in keeping customers satisfied and safe from fraud As banks FinTechs and TPPs work together to deliver new services for their customers they must also stay vigilant against fraudsters In addition their joint anti-fraud efforts will require a renewed and deeper understanding of how fraud works in the digital era Changing the way FIs perceive fraud will be key to responding to incidents when they are detected and preventing future attacks The following Deep Dive examines how cybercriminals are getting more creative in the digital era why FIs should use genetic mapping as a fraud-fighting strategy and how new players can establish themselves as legitimate partners in the fight against fraud A Wide Array of Options for Fraudsters Digital bankings growth is presenting bad actors with new ways to obtain consumers data fraudulently perform financial transactions and steal money There have recently been several notable events that highlight just how creative these fraudsters have become A gang of cybercriminals in 2016 used malware and the international SWIFT system to make roughly 35 withdrawal requests from Bangladesh Bank in an attempt to steal roughly 1 billion Thirty of those transactions valued at approximately 850 million were blocked by the Federal Reserve Bank of New York The thieves were still able to transfer about 101 million to banks in Sri Lanka and the Philippines before they were flagged because of a spelling error Only 20 million was recovered Tesco Bank which is owned by a supermarket group in the UK was also hit by cybercriminals that year Fraudsters were able to withdraw 25 million 322 million from approximately 20 000 accounts Analysts believe the hackers exploited a vulnerability in Tesco Banks website that appeared when customers made online purchases Cybercriminals are becoming more brazen and banks are struggling to stay on top of their anti-fraud efforts Just 37 percent of banks had completed fraud detection management systems or were transforming them as of late 2018 and only 16 percent were able to detect fraud as it happened AI and ML solutions are promising to boost that portion by stopping fraudulent incidents as they unfold These tools can also be used to prevent potentially illegal acts Before this can happen financial ecosystem players must understand how fraud is committed and how it will change in the future Decoding the Genetics of Fraud Embedded in the financial ecosystem are FIs and consumers who are using financial tools like credit cards debit cards and bank accounts to perform transactions Nestled among these trustworthy players are the fraudsters who stay active in the broader payments ecosystem The complexity of the payments ecosystem can be compared to DNA and genomes when it comes to detecting bad actors and highlighting how their activity differs from acceptable behaviors By using AI ML and risk analysis fraud can be pinpointed and ideally prevented This DNA-like approach to fighting fraud creates a visual analysis of illegal activity and highlights different connections such as the machines used to commit fraudulent acts or the locations the crimes were committed ultimately uncovering the layers of fraud from individuals to bots to broader networks This approach can help to quickly detect it when aided by tools like AI and ML The effects of financial fraud can be far-reaching including lost funds for consumers and damaged reputations for FIs A biology-based approach to detecting and isolating illicit financial activities could keep cybercriminals harmful impacts from spreading as they get more creative Other Tools to Fight Fraud AI and ML are valuable fraud-fighting assets but human beings also play a central role Thats why a poor culture of control within banks could be an advantage for fraudsters A lack of control culture in which employees fail to follow the correct processes to address fraud is highly enticing for fraudsters targeting FIs A recent report stated that 72 percent of European fraudsters found a companys weak control practices to be a significant opportunity Because fraudsters want to exploit internal vulnerabilities the best way for an FI to improve its anti-fraud measures is to think human and implement a strong anti-fraud culture that works alongside AI and ML technologies Fraud is continuing to evolve in the age of digital banking Examining suspicious activity and attempting to understand fraud at its root level combined with support from human analysts could be key to staying one step ahead of criminals LATEST INSIGHTS Our data and analytics team has developed a number of creative methodologies and frameworks that measure and benchmark the innovation thats reshaping the payments and commerce ecosystem Check out the latest PYMNTS Faster Payments Tracker AI APP Fraud deep dive Digital Banking Digital Banking Tracker DNA Feedzai fraud Human Insight machine learning News PSD2 SWIFT