• | 9:00 am

Old technologies in the Middle East pose new financial crime risks. Will AI be the solution?

As concerns over the insidious threats of money laundering and fraud continue to loom ahead, here’s how a new wave of companies in the region are wielding AI to combat them

Old technologies in the Middle East pose new financial crime risks. Will AI be the solution?
[Source photo: Anvita Gupta/Fast Company Middle East]

This January, a nefarious storm brewed within the local banks of Kuwait, leaving dozens of unsuspecting clients becoming victims of a calculated attack by hackers. They utilized a new weapon—emails with cunningly crafted links to grant remote access to the victims’ smart devices to plunder the banking data realm.

As the news broke, the Cybercrime Investigation Unit sprang into action and revealed a disturbing pattern—each financial fraud operation originated from fraudulent emails, expertly camouflaged to resemble official communications. These fraudulent messages bore the logos of trusted entities, including the Ministry of Communications and courier companies like DHL and Aramex.

The emails would begin innocently, tempting recipients with the promise of a long-awaited delivery for a seemingly small price of $5.04. Once the victims acquiesced, the theft unfurled, far surpassing the meager fee. The stolen amounts ranged from a relatively smaller sum, like $1,010, to $50,5446 per operation. 

In this relentless battle against cybercrime, one crucial truth emerged—the sanctity of personal information could no longer be taken for granted. 


To up their fight against increasing financial crimes, companies in the Middle East are wielding AI to combat the insidious threats of money laundering and fraud.

Mozn, a Saudi-based AI tech company recently expanding in the UAE, leverages AI, pattern recognition, and advanced algorithms to combat fraud, money laundering, and financial crimes. FOCAL, Mozn’s AI-powered risk and compliance platform, exposes an application programming interface (API) to allow its clients to integrate anti-money laundering (AML) and anti-fraud compliance capabilities into their core systems. 

Dr. Mohammed Al Hussein, CEO and Founder of Mozn, notes that FOCAL combines data points to score risk for fraud and money laundering “that automates a custom next action based on the configurable rules and the organization’s risk appetite.”

“It can detect user behaviors such as end-user location, if the beneficiary is within a specific country or region and user network, if the beneficiary is part of a suspicious network or fraudulent ring, among others,” he adds.

According to a recent global anti-money laundering research conducted, it was revealed that 57% of institutions have already incorporated or are about to integrate AI and machine learning into their AML compliance department. The study further highlighted that one-third of financial institutions are expediting their adoption of AI and ML for AML technology. 

According to Grozdana Maric, Head of Fraud & Security Intelligence EMEA Emerging and AP, SAS, “the findings highlight the growing recognition of AI and ML as crucial tools in combating money laundering, with financial institutions prioritizing enhanced investigations, streamlined filings, and operational efficiency, even during the pandemic.”

As for SAS, the company adopts a hybrid analytics approach to detection embedded in the company’s technology – combining AI, ML strategies, and network analytics, which helps financial institutions easily identify complex threats. A powerful method leveraged, Maric points out, for identifying the same identities, especially in cases where criminals manipulate certain data elements, is Entity Resolution, which enables the company to recognize user data and connect related entities within the data. 

Additionally, she signals that these data analysis and detection processes are enhanced by leveraging text mining. “Extracting valuable information from unstructured text sources, such as SWIFT messages or customer complaints, helps us enrich your dataset and gain valuable insights.”

According to a 2018 PwC report, the Middle East AI market is projected to reach a staggering value of $320 billion by 2030. Realizing the immense economic potential of AI in the Middle East, companies are increasingly tapping into the space. A case in point is a mobile banking services provider that safeguards customers against payment fraud, identity theft, and phishing attacks through AI-powered fraud management systems. 

Last December, NOW Money formed a strategic partnership with another player in the space, ThetaRay, an Israel-based provider of AI-powered transaction monitoring technology, to enhance its users’ financial security and stay a step ahead of financial crime by uncovering both known and unknown money-laundering crimes, as well as preventing attempts to violate sanctions in real-time.

Leveraging ThetaRay’s cutting-edge AI-based SaaS AML, NOW Money has access to its big data analysis and proprietary algorithm science expertise. “This collaboration has further empowered us to detect anomalies accurately while minimizing false positives,” says Noel Connolly, CEO of NOW Money. 


Traditional financial crime detection methods like rule-based systems and manual reviews face many challenges. One of the most daunting is the enormous volume of financial transactions happening daily across the globe.

Dr. Al Hussein observes that the manual review of each transaction is nearly impossible, and rule-based systems often struggle to cope with the intricate complexity and variety involved.

“Rule-based systems that flag transactions based on predetermined rules also often generate a high number of false positives, which means they flag transactions as suspicious that are legitimate, leading to wasted time and resources on investigating these false leads.” 

Furthermore, financial institutions face significant challenges in meeting customer expectations, complying with evolving AML, countering the financing of terrorism regulations, and combating the pervasive threat of financial crime. These challenges, Maric notes, are exacerbated by the presence of siloed systems and fragmented data sources. “To effectively mitigate fraud and AML risks, a comprehensive view of customer activity, fund sources, and ownership structures is crucial.”

Today more than ever, companies are coming to terms with the fact that addressing costly regulatory compliances, managing the exponential growth of data volumes, and deploying advanced technologies cannot be accomplished through manual operations alone.

“Automation plays a crucial role by allowing companies to reduce the need for repetitive tasks such as data consolidation, cleansing, coding, and cross-system data analysis,” says Bassem Awada, General Manager – MENA & Vice President of Global – Key Partnerships at global payments firm, TerraPay. 

In a significant crackdown in March this year, Saudi authorities revealed the capture of a Saudi citizen and an Arab national involved in approximately 150 financial fraud cases. The duo orchestrated their crimes by impersonating government officials and entities, ultimately stealing an estimated $2.93 million. 

Connolly remarks that AI’s importance in fraud detection cannot be overstated as it offers robust and comprehensive data sets for unmatched detection capabilities. “AI’s continuous development, combined with comprehensive data sets, ensures robust protection against various fraudulent activities.”

In the ever-evolving world of financial crimes, it’s a significant challenge for traditional approaches to keep pace with financial criminals’ adaptability. Dr. Al Hussein warns that these conventional methods often falter when identifying nuanced patterns and intricate links within and across transactions, especially when confronted with vast and complex datasets. “They also may not integrate various types of data effectively like transaction data, customer information, or external risk indicators.”

Non-integrated systems remain other impediments, given their essence of hindering information sharing, thereby making it difficult to identify suspicious activities or money laundering schemes. Maric opines that integrated solutions that provide a holistic view of customer data must address these challenges. “By breaking down data silos, institutions can better understand customer behavior, transaction patterns, and associated risks, thereby enhancing their ability to detect and mitigate financial crime risks.”


Experts allude that AI’s usage requires continuous development and updates to adapt to evolving data and emerging fraudulent schemes. As Yahya Alazri, Security Consultant Ministry of Transport, Communication & Information Technology, says, “When we talk about AI in any application, we talk about learning.”

Alazri contends that as learning represents a significant constraint in AI applications, the effectiveness of AI is contingent upon the quality of its learning model. “If we implement a robust learning model, we will enhance the security of our applications. Conversely, a weak model will result in numerous weaknesses and vulnerabilities within the system.”

It was recently revealed that in the run-up to the World Cup in Qatar, email-based phishing attacks targeting the Middle East witnessed a twofold increase in October, according to research conducted by Trellix. These phishing emails claimed to originate from the FIFA help desk, ticketing office, specific team managers, or departments.

Connolly hints that continuous machine learning is imperative for AI to remain effective in a dynamic world characterized by evolving societal trends and habits. “Neglecting updates or using inappropriate data sets can create vulnerable spots where users are susceptible to specific scams.”

Even as money laundering continues to plague the world of payments, one major challenge identified by experts is that unstructured data from multiple sources, manual rule-building, and complex analytics routes confound the financial system’s ability to scale across borders. 

According to Awada, one answer to this slow-moving, inefficient process can be found in new-age AML detection systems, given that advanced AI algorithms can analyze global transaction patterns, monitor cross-border flows, and identify suspicious activities indicative of fraud. “In the context of the Middle East’s prominence in cross-border transactions, AI will play a vital role in enhancing cross-border fraud detection.” 

Looking back, it is evident that fraud and financial crimes have not disappeared but evolved, changing their methods, channels, and tactics. This global trend, Maric contends, is equally relevant for the Middle East region. She highlights that money laundering and other financial crimes escalate worldwide, accompanied by increasingly sophisticated techniques to evade detection. 

To effectively tackle these challenges, the way forward for banks is to combine machine learning and advanced algorithms like random forest, gradient boosting, and deep learning. “There is a growing adoption of Robotic Process Automation to automate manual processes such as fetching third-party data or escalating alerts, further enhancing efficiency in AML operations,” says Maric.

The path toward a secure and resilient future in financial crime prevention relies on embracing AI and its ability to adapt, learn, and mitigate risks in an increasingly complex and interconnected world. 

As Dr. Al Hussein aptly says, through “harnessing the boundless powers of AI, we can steadily and methodically eradicate risks, one click at a time.”

  Be in the Know. Subscribe to our Newsletters.


Misbaah Mansuri is a UAE-based senior freelance writer who particularly loves covering topics at the intersection of technology and culture. Her work has been featured in the likes of BBC, National Geographic, and Digital Studio Middle East, among other leading publications. Gaming and technology for good spark her curiosity. More

More Top Stories: