AI on the frontline in Iberpay's fight against financial fraud

| minute read

Spanish payment system Iberpay adds AI and machine learning to its Payguard anti-fraud service to provide unprecedented cross-sector insight for banks to help crack down on financial crime. 

 

How Machine learning is transforming cyber crime detection

Artificial Intelligence and Machine Learning have dramatically enhanced our Payguard anti-fraud service. This powers up Europe’s banking industry to more effectively crack down on financial crime.

Iberpay manages the Spanish payment system, a critical element of Europe’s interbank infrastructure and payments network. We connect all banks operating in Spain and Andorra to process account-to-account payments for individuals, businesses, and governments in the SEPA area.
Last year Iberpay processed over 3 billion transactions worth 2.7 trillion euros.
Instant payments make up 52% of all credit transfers in Spain – more than the 19% average in the rest of Europe.

Instant payments are fast and convenient for our customers, but also criminals. So, we created Payguard, a complete service including 6 tools to prevent, detect, and anticipate fraud in these transactions.
AI offers an opportunity to strengthen our Payguard service.
We use AI and machine learning to proactively analyse transactions and identify suspicious patterns, fund movements, and transfer timelines.
Iberpay and Sopra Steria worked together to develop the algorithms, deploy the solution, migrate to the cloud, and roll out a sophisticated analytics platform to track data analysis.

The first version of the solution was developed in just six months.
It provides unprecedented cross-sector insight for banks, helping them crack down harder on financial crime.
While other tools are deployed on-premise, the AI and machine learning elements are hosted in the cloud, better suited to storing and processing large data volumes.
The algorithms automatically monitor transactions, cross-checking against known fraud indicators and highlighting suspicious activity.
Banks frequently provide feedback to the analytical engine to confirm whether flagged activity is indeed fraud, meaning the AI is always learning and constantly improving future detection.
Next, we want to keep evolving Payguard, adding new algorithms to further reinforce our commitment to stopping cybercriminals in their tracks.

Iberpay is based in Madrid. The company manages the Spanish payment system (SNCE), connecting more than 140 banks across Spain and Andorra to the rest of Europe. It also manages the cash distribution system (SDA) in Spain. 

The company processes account-to-account payments for individuals, businesses and governments in the region. In 2023 it processed more than 3 billion transactions - worth 2.7 trillion euros - with an average of 12 million transactions a day. 

Iberpay has provided the interbank services for processing SEPA (Single Euro Payments Area) instant transfers since 2017, processing an average of 2.7 million instant transfers daily and 3.9 million at peak times.  

Instant payments are more heavily established in Spain than in the rest of Europe, accounting for more than half (57.1%) of all credit transfers in the country, compared to the 19.7% average in the rest of Europe.  

Enhancing fraud detection and preventing money laundering 

Instant payments are highly valuable to customers also make them attractive for criminals. Iberpay’s central role in facilitating instant payments means that continual improvements in fraud prevention are critical to protect customers’ cash and prevent money laundering.  

Iberpay developed Payguard, a collection of five tools that work together to help identify, prevent and anticipate fraud. Payguard is the result of a joint collaboration of the Spanish banking community that are connected to Iberpay’s account-to-account payment service.  

Identifying strange behaviours is essential in recognising potential fraud but can be time-consuming and challenging, especially as fraudulent practices evolve. Iberpay wanted to strengthen the analytical powers of Payguard to recognise unusual patterns that could indicate fraud, as well as building its warning signs data to improve future fraud detection. 

Incorporating AI to improve detection  

Sopra Steria worked with Iberpay to develop and add AI and machine learning algorithms to Payguard. These help to identify abnormal occurrences, such as suspicious patterns and unusual fund movements and transfer timelines.  

The algorithms automatically monitor transactions and cross check against known fraud indicators. Any suspicious activity is flagged for entities for further checks.  

Sopra Steria migrated the AI and machine learning tools to the cloud. The company also developed and deployed an advanced data analytics platform to track the data analysis.  

A variety of use cases were also developed to improve the detection of potential fraud and money laundering in bank transfers. By detecting such behaviours banks have a wealth of cross-sector information to analyse to further improve their fraud prevention measures and the AI is able to continually learn and improve.  

Artificial Intelligence and Machine Learning, seamlessly integrated into our collaborative solution, have significantly enhanced the effectiveness of PayGuard, our anti-fraud service.  

Juan Luis Encinas

CEO, Iberpay

Better detection of financial crime, enhanced fraud-prevention learning and improved efficiency of instant payments 

The first version of the solution was developed in only six months. The incorporation of AI and machine learning algorithms within Payguard have dramatically enhanced Iberpay’s fight against fraud. It has enabled improved detection of financial crime, increased the efficiency of instant transfers and provided valuable cross-sector information for banks to analyse.  

Regular feedback from banks about whether flagged activity is fraud enables the AI to constantly learn, enabling improved future detection of fraud.  

Next steps will see the two companies work together to enable real-time monitoring, allowing transactions to be analysed and identified as fraud as it happens and increasing the ability to eliminate it at source. 

  • 3 billion Total number of transactions processed by Iberpay and analysed by Payguard
  • + 140 banks Payguard analyses account-to-account transaction transactions from over 140 entities
  • 6 Months taken to develop the first version of the solution
Continual evolution of preventative measures, and embracing tools such as AI and machine learning, are crucial to fighting financial fraud.  

Miguel Fernandez Sanz

Financial Services Consulting Lead, Sopra Steria Spain

artificial-intelligence

banking-solutions

cybersecurity

data

digital-transformation

innovation

Related content

ADEO reshapes its Data Platform with Cloud-based technologies

To sustain its strong growth, the company has laid the extension of its business on a Platform and Data strategy. To propel this ambitious plan, ADEO has decided to migrate its entire IT system into Google Cloud, applying a datacenterless strategy.

Proposed acquisition of Fidor Solutions, the software subsidiary and digital banking specialist of next-generation bank Fidor Bank

Sopra Steria recognized as a Leader in Agile Development & DevOps Services by global analyst firm NelsonHall

NelsonHall’s NEAT vendor evaluation reflects Sopra Steria’s overall ability to meet future client requirements as well as delivering immediate benefits to Agile Development & DevOps Services clients. This evaluation assessed all the major providers in this segment worldwide.