Sopra Steria, the technology services giant, has worked with the Glasgow-based Financial Regulation Innovation Lab (FRIL) in a bid to harness AI in the fight against economic crime.
The consulting, digital services, and software development firm has produced an ‘actionable’ whitepaper which looks to help the financial services industry use the technology to better protect against the rise of sophisticated online threats.
They include deepfake technologies, phishing and spear phishing, automated social engineering, credential stuffing, synthetic identity fraud and others.
In a detailed document spanning 20 pages, the organisation has produced a wide-ranging set of recommendations and mitigations through the use of AI-driven software to enhance the suite of protections against financial criminals.
It proposes integrating AI technology to offer a ‘holistic and dynamically adaptable approach to financial crime detection and prevention’.
‘By implementing this framework, financial institutions can significantly enhance their capabilities to identify, mitigate, and prevent financial crimes, ensuring a more secure financial ecosystem,’ the paper, produced in collaboration with the University of Strathclyde, argues.
Economic crime, including fraud and scams, continues to rise, driven by technological advances and economic factors such as the cost-of-living crisis, the document states.
“Fraud accounted for 40 per cent of all crimes in 2023, with costs to society exceeding £6.8 billion. Scams exploit consumer vulnerabilities, ranging from online purchases and investment schemes to romance scams and blackmail. Money laundering is also pervasive, with estimates suggesting it accounts for 2-5 per cent of global GDP annually.
“This situation is only being made worse due to the cost-of-living crisis, with consumers more willing to take risks, whether that be making ‘too good to be true’ purchases online, accepting opportunities that offer quick payouts, or romance scams resulting from individuals feeling alone and isolated.
“Mix this with the rapid advancement and accessibility of technology, in particular Generative AI, and scams are becoming more realistic and security layers becoming easier to breach when in the hands of bad actors. For example, it may be exciting to create fake images of yourself with interesting backgrounds using AI, but in eyes of the fraudsters, that is fake IDs, synthetic documents and an opportunity to create fake bank accounts.”
While AI-enabled systems are effective at identifying potentially criminal transactions, they often generate large sets of false positives – instances where legitimate transactions are incorrectly flagged as suspicious. False positives can result in delays, inconvenience, and additional scrutiny for customers.
However, Sopra Steria has devised a methodology to reduce false positives.
It has developed an innovative tool that constructs ‘optimal hybrid business rules with AI at the core’, sitting as a layer on top of existing fraud solutions. The rules delivered by the tool daily allow firms to be in full control of their fraud management environment by alerting only the transactions that need to be, balancing their risk appetite with resource management, whilst ensuring their customers are protected, all in a way they have not been able to before.
“Fundamentally, the solution takes in historical transaction data and outputs a set of ‘rulesets’, optimised for three objectives, minimising false positives, maximising true positives and maximising detected fraud value,” the paper states.
Going forward, it hopes the fully automated and self-learning model will lead to a ‘hands-off approach’ until the resulting rulesets are output. ‘This means the only time required from the Data Scientists or Fraud Analysts is for reviewing and selecting the appropriate ruleset’, the paper notes, highlighting the fact that the solution is to retain ‘humans in the loop’.
“The rapid advancement of AI technologies presents both significant challenges and remarkable opportunities in the realm of financial crime detection. As criminals become increasingly proficient at exploiting these technologies, the imperative for financial institutions and consultancy firms to adopt advanced AI-driven solutions becomes ever more critical. This white paper has highlighted the diverse range of AI approaches currently employed by criminals and the corresponding AI techniques that can effectively counteract these threats. By leveraging Machine Learning for anomaly detection, financial institutions can build a resilient defence against financial crimes.
“The proposed end-to-end framework represents a comprehensive solution that not only addresses current threats but is also adaptable to future challenges. Implementing such a framework will enable financial institutions to stay ahead of criminal tactics, ensuring a robust and secure financial environment. In conclusion, the integration of advanced AI technologies is not just a strategic advantage but a necessity for the sustainable protection of financial assets and customers.”
To download the whitepaper, visit here.