Artificial intelligence – the new weapon detecting financial fraud

Artificial intelligence – the new weapon detecting financial fraud


Fraudulent activity is a major risk faced by financial institutions such as banks and insurance companies around the globe. They together spend billions on developing anti-fraud technologies. Artificial intelligence (AI) has emerged as the ultimate weapon in the fight against financial fraud. The Swiss start-up NetGuardians is one of the players using the latest AI technology to detect fraud.

Fraud rates remain at record highs. Nearly half of the 5,000 companies polled by the consultancy PwC responded that they had suffered from fraud over the past 2 years. Fraudulent activities such as hacking attacks, scamming, internal fraud, cost these polled companies alone more than 42 billion US dollars, according to PwC’s Global Economic Crime and Fraud Survey 2020. A small proportion of these losses are ever recovered. It is here that the solutions developed by companies such as NetGuardians come in handy.

The first line of defense against fraud in place such as application firewalls, two-factor authentication systems... works well. “The weakness targeted by fraudsters is the human link, which is much easier to trick. A single phone call can do the trick and cause massive losses for a company or individual,” explains Joël Winteregg, the CEO and co-founder of the fintech start-up NetGuardians.

The Yverdon-les-Bains-based company has developed an AI propriety technology that analyses financial transactions in three steps: anomaly detection, fraud recognition and adaptive feedback. “There are billions of financial transactions carried out by our clients every year, but thankfully very few cases of fraud among these. Our software, however, is able to immediately detect these exceptional, fraudulent payments,” Winteregg adds. NetGuardians counts more than 60 banks and wealth managers across the globe among its clients.


Booming data protection market

The global data protection market, valued at 15 billion US dollars in 2020, is forecast to grow by more than 30 percent between 2020 and 2027 annually, according to Grand View Research. The market is dominated by large players in terms of revenues, but in terms of service providers the players offering fraud detection services are smaller actors. NetGuardians competitors, such as BlockFraud, SEON and identiq, are mainly based in the US, UK and Israel. Geneva-based Temenos also offers AI-powered fraud prevention solutions enabling clients to detect fraud and mitigate reputational risk.


Great potential for collective AI learning

The next step within AI in the financial sector is the sharing of non-confidential client information between banks. Switzerland has a competitive advantage here, as it benefits from the presence of a large number of banks on its territory. The access to large volumes of real data is thus easily within reach, a real goldmine with regard to AI. The focus is not on known fraudulent IBAN (bank accounts) numbers or computer IP addresses but on sharing genuine payment behavior in an anonymous manner.

“To combat fraud, you need access to real data, not to test data. The latter can turn out to be completely useless,” underlines Winteregg. “The algorithm needs to be able to detect genuine accounts and payments. A large one-off payment to an unknown beneficiary in Singapore for example typically sets off alerts.” Access to collective AI data would avoid this type of false alarms. The great advantage of collective AI is that is does not require any Customer Identification Data (CID) being shared. “We will move forward in this area in the coming months. This will benefit the entire Swiss banking sector. Progress in this area will ultimately lead to enhanced protection against fraud for Swiss banks – in the end benefiting all honest individuals and companies,” says Winteregg.


Joel Winteregg, CEO at NetGuardians