An epic battle is taking place in the world of banking fraud and financial crime. Banks and fraudsters are using the latest technology, and the winner of the war has yet to be determined.
While AI and GenAI hold incredible transformative potential for banks and are being used in countless ways to shape the future of banking, they also present a double-edged sword. Banks are using AI and GenAI in their efforts to combat fraud; however, fraudsters are also using these advanced technologies to commit crimes. At the moment, the fraudsters are winning the battle. It’s up to banks to be adaptable, agile and creative if they hope to win the war.
Fakes are easy to create – and tough to catch
As AI advances, criminals are increasingly using it to perpetrate fraud.
According to Iain Brown, PhD, Head of Data Science at SAS in Northern Europe, “Generative AI and LLMs, despite their transformative potential, harbor significant fraud and security risks.”
GenAI is relatively inexpensive for fraudsters, but beyond that, it also enhances their capabilities by allowing them to develop fraud scams, analyze their effectiveness and continuously update methods to evade detection systems. GenAI can be used to create deepfake videos, voice clones and other biometric alterations that further help bad actors breach bank systems, making it harder for banks to keep up as the frequency and sophistication of incidents increase. AI is helping malicious actors revive scams that banks believed were long gone, like check fraud, while also expanding phishing scams.
They’re using AI to sift through data to identify potential victims and tailor scams to exploit their vulnerabilities more effectively. They’re also deploying chatbots to engage with victims and manipulate them into paying inordinate sums. These chatbots are particularly effective because they are easy to scale and replicate, and they require minimal human involvement. GenAI makes it easy for fraudsters to create convincing emails, messages and images that deceive victims and are increasingly difficult to spot. With this surge in fraudulent activity, what’s next?
A rising tide lifts all boats
While the rising tide analogy is often used in a positive light, this particular rising tide isn’t playing out well for banks. Fraudsters have a growing armory of AI technologies at their disposal, and they’re intent on sinking the ship.
Fraud losses continue to rise and are becoming more difficult to calculate due to the wide range of hard and soft costs banks face – including direct financial losses, investigation expenses, legal fees, compliance costs, reputational damage and operational disruption, among others. Some estimates put global fraud losses at an astounding $5 trillion per year, and Deloitte estimates that Gen AI alone could cause fraud losses to reach more than $40 billion in the US by 2027. That’s up from $12.3 billion in 2023 – a compound annual growth rate of 32%.
In an increasingly competitive market, there’s another significant loss that’s top of mind – customers. Globally, more than 67% of customers say they would switch to another provider if they experienced fraud or felt that another company provided better fraud protection.
These are tides that rise and then drown.
As we enter 2025, we see global banking contending with the double-edged sword that is AI. Malicious actors are deploying AI faster than banks can adopt, creating more sophisticated scams to circumvent banks' traditional security measures. However, on the other edge, banks are heavily investing in AI-powered fraud solutions that learn and adapt in real time. Though the challenges of realizing production value from AI initiatives persist, the opportunity is real for those who address fraud proactively and expediently, improve the customer experience and deepen the trusted relationships with their customers. Stu Bradley, SVP and Global Head of Risk, Fraud and Compliance Solutions, SAS
A multi-pronged weapon
For banks to win the war, they cannot use a singular approach but instead must consider a multi-pronged strategy. Iain Brown also noted, “A proactive, multi-faceted approach encompassing technology, regulation, and education can help mitigate these risks, ensuring trustworthy AI development and usage. By maintaining vigilance and fostering collaboration and innovation, we can harness the advantages of GenAI while safeguarding our digital ecosystems from the ever-adapting threats posed by malicious actors.”
In the fight against AI-driven fraud and financial crime, banks must focus on these critical areas to stay ahead of evolving threats.
- Technology: While banks have long used AI and embedded it in their fraud detection systems, ongoing efforts are needed to evolve these detection and prevention systems continually—and data is central to these efforts. As banks work to continually improve their fraud systems, incorporating more data sets (both structured and unstructured) can help train models to better identify subtle patterns and indicators of fraud, ultimately transforming monitoring capabilities.
- Regulation: AI regulation around the world is fragmented. For the global community to make progress against a common adversary, a more aligned regulatory approach and faster action are crucial. This collaborative approach would not only allow banks and other organizations to operate more smoothly across regions, but it would also strengthen partnerships with governments and law enforcement agencies working to shut down fraud and financial crime networks.
- Education: The proliferation of fraud and the ways fraudsters target individuals call for a better approach to education. This battle is no longer just the bank’s responsibility – it involves everyone. Those with the knowledge must share it and train others. It’s also in banks’ best interest to invest more in educating their customers on how to better identify fraud – especially as AI fuels the creation of deceptions that make it increasingly difficult to distinguish between the real and the unreal.
When the war is won
Is it truly possible to win this war?
History has shown us that parts of humanity will always find not only the will but also the ways and means by which to plunder. Where there is money to be made, there will always be those who would prefer to take it rather than earn it. As we continue the fight against fraud and financial crime, perhaps human ingenuity and creativity, combined with the vast potential of AI, can eventually prevail against the bad guys.
