Scam City: The Rise of AI-Driven Financial Fraud and the Fight Back
- By Winston Thomas
- May 26, 2024
A disturbing evolution is underway in the digital world's shadowy corners. Once a domain of opportunistic swindlers, financial crime is now a playground for tech-savvy masterminds.
But the weaponization of generative AI and synthetic data is tipping the scales, turning financial fraud into a formidable beast.
Imagine AI-powered chatbots impersonating your bank manager, persuading out sensitive information with unnerving ease. Or hyper-realistic deepfakes—videos and audio so convincing they could fool even the most discerning eye—become tools of deception, manipulating victims into parting with their hard-earned cash. This isn't a scene from a sci-fi thriller; it's the chilling reality of modern financial crime.
But the battle isn't lost. Data analytics, the modern-day Sherlock Holmes, is stepping up to the challenge. And companies like SAS are using a combination of time-tested data analytics techniques and AI to fight these AI-driven scams.
Tales from the trenches: AI's arsenal of scams
Talking on the sidelines of the SAS Innovate Singapore, Keith Swanson, director of fraud and security intelligence at SAS, highlights three key trends in financial crime: authorized push payments (APP), account takeover and identity theft, and mule accounts.
APP involves tricking individuals into making payments to fraudulent entities. Account takeover and identity theft occur when criminals use stolen information to gain access to accounts or open new ones. Mule accounts funnel illicit funds, often through a network of seemingly dormant accounts that suddenly see large cash inflows.
These trends are exacerbated by scammers' increasing use of synthetic data and AI. Synthetic data, artificially generated information that mirrors real data, creates convincing fake identities and transactions. AI automates the creation of synthetic data and identifies vulnerabilities in financial systems. It's a technological arms race, making the role of data analytics even more critical.
Unmasking the deception with the AI-based Sherlock Holmes
SAS sees the vast ocean of information as a treasure trove of clues that can expose fraudulent activity. It’s like a high-stakes game of hide-and-seek, where criminals try to blend in, but their digital footprints betray them.
SAS combines predictive analytics and model-based scenarios like a digital magnifying glass to explore these footprints. These meticulously scan mountains of data to quickly uncover hidden patterns and anomalies that traditional rule-based techniques overlook—neatly displayed in financial crime risk dashboards. It's like a scene straight out of a crime thriller, where a seemingly innocuous transaction suddenly triggers a red flag, alerting investigators to a potential fraud in progress.
Swanson advocates companies to “layer” more techniques. He views it like panning for gold: the better the mesh (or filters in this case), the better your chance of discovering a valuable “nugget” of scam information. He also highlights significant improvements in reducing false positives—a menace for financial crime busters that seems to be wasting time going after dead ends—using dynamic segmentation.
But data analytics isn't just about detection; it's about prevention. SAS's real-time monitoring capabilities act as a digital tripwire, halting suspicious transactions before they wreak havoc. It's like having a security guard stationed at every digital doorway, scrutinizing every entry and exit.
Collaboration becomes pivotal against financial crime
Swanson says collaboration is critical in this high-tech game of cat and mouse. SAS solutions are engineered for this area. For example, SAS's end-to-end fraud detection and prevention solution marries multiple channels and lines of business on a single platform.
The solution leverages SAS Viya and allows fraud and anti-money laundering (AML) teams—two teams that usually work separately—to join forces and share insights. It empowers what the industry calls the FRAML (fraud and anti-money laundering) strategy, creating a digital war room where the good guys from these departments strategize, pool their resources, and outsmart the enemy.
As Swanson notes, a holistic approach to financial crime is becoming increasingly pivotal as it becomes more complex and criminals share more information among themselves. By working together, different teams involved in financial crime can leverage their collective knowledge and expertise to combat financial crime more effectively.
The scam battleground also extends beyond the confines of individual institutions. It's a global war that requires collaboration between financial institutions, telecommunication companies, and law enforcement agencies, emphasizes Swanson. It's about sharing data, pooling resources, and presenting a united front against fraud. A platform that easily integrates with third-party information and data becomes a major gamechanger.
Outsmarting the scammers: AI-Powered countermeasures
As scammers become more AI-savvy, so must the tools used to combat them. Enter AI-powered analytics, the digital bloodhounds with an uncanny ability to sniff out fraud.
SAS already uses embedded machine learning methods as tireless detectives, working around the clock to identify patterns that would elude even the most seasoned human investigator. They can learn from past decisions and help financial crime teams prioritize and speed up investigations. Entity resolution, pattern matching and network detection techniques further unravel unique entities and hidden risks within complex financial crime patterns.
As Swanson puts it, "SAS is very focused on…demonstrating the viability of AI" in combating scams. In the ongoing battle against financial crime, understanding and harnessing the power of data analytics is not just a competitive edge; it's a necessity.
But AI isn't just a watchdog; it can also be a weapon. SAS's platform streamlines the investigative process, enabling law enforcement agencies—the most challenging part in combating scams as many agencies work in siloed ways—to move swiftly and decisively.
While AI is important, Swanson believes there is no substitute for human intuition and expertise. SAS's "human-in-the-loop" approach ensures that the final decisions rest with human investigators, adding a layer of critical thinking to the equation.
Fighting for the soul of FSI
The stakes are high in the digital badlands of financial crime. The fusion of AI and financial fraud has created a formidable foe that threatens the foundation of trust upon which our financial systems are built. But as the criminals evolve, so too must our defenses.
SAS and other innovators are in charge of harnessing the power of AI and data analytics to forge a new path forward. This is not just a war—it's an arms race for the future of finance.
Image credit: iStockphoto/Andreus
Winston Thomas
Winston Thomas is the editor-in-chief of CDOTrends. He likes to piece together the weird and wondering tech puzzle for readers and identify groundbreaking business models led by tech while waiting for the singularity.