Can artificial intelligence (AI) help combat money laundering? Spanish business Lynx Tech is embracing the surging technology to fight against fraud and financial crime, with the company securing 76 billion global transactions from 330 million users in the last year. The Spanish company, founded thirty years ago by experts at the Autonomous University of Madrid, has presented an AML detection solution (Anti-Money Laundering, in Spanish) designed to assist financial institutions in identifying high-risk individuals and entities with precision and speed.

Over the past few decades, models for combating money laundering and financial fraud have undergone a profound transformation from rigid, rules-based systems to agile, adaptive AI-driven technologies. Lynx Tech explains that legacy solutions once relied heavily on static rule sets and supervised models trained on historical data, which, while effective against known threats, struggled to keep pace with fast-evolving criminal tactics. This limitation often led to high false positive rates, overwhelming analysts and frustrating legitimate customers.

Moreover, the increase in global sanctions and the evolution of sanction evasion tactics have made traditional detection methods slow and prone to false positives, making it necessary to use newer, more effective methods. Furthermore, the United Nations (UN) estimates that between 2% and 5% of global GDP is associated with money laundering each year, between 715 billion and 1.87 trillion euros. Lynx Tech, for instance, announces that it has already saved US$1.6 billion for its clients. In addition, the company helped a European financial institution achieve a 700% improvement in money mule detection rate and a 35% increase in scam detection.

Fighting money laundering with precision

The Lynx Tech AML detection solution is powered, as already mentioned, by AI, is highly configurable, and is designed to evolve with the changing regulatory landscape, the company explains. Specifically, it supports more than 100 languages, handles phonetic and spelling variations, offers adaptable watchlist management, and uses proprietary Natural Language Processing and machine learning models to improve name similarity scoring.

According to the business, the solution is designed to analyse hundreds of transactions per second, with an average response time of less than one second and a false positive rate below 1%. “Configurable solutions allow fintechs to tailor risk-based approaches, applying higher thresholds for lower-risk products while maintaining the agility to adjust quickly if risks evolve,” the company explains.

Latest AML screening solution

In March, Lynx Tech introduced its most advanced AML screening solution, which utilises its proprietary natural language processing models and machine learning to boost its capabilities. Moreover, the company noted that it allows customers to customise rescreening rules according to their policies. The company also said its AI solution offers a scalable infrastructure to accommodate different volumes of transactions, cloud-native scalability for rapid and global implementation and real-time insights. “The landscape of financial crime is shifting rapidly, and financial institutions need a solution that not only keeps up, but gets ahead,” underscored Dan Dica, CEO of Lynx Tech.