Snapping at their heels are new-age fraudsters who are increasingly deploying sophisticated tactics to deceive and find loopholes in the system that cost businesses and individuals millions each year.
Combating payment fraud and minimising its potentially severe financial and reputational impact has, therefore, become a critical priority for businesses worldwide.
In Ireland, the total value of fraudulent payments climbed by 26% in 2023, rising to €126 million from €100 million in 2022, according to a new research by the Central Bank of Ireland. Among payment methods, card fraud accounted for the highest rate, with 0.034% card payments by value being fraudulent. In contrast, the European Economic Area suffered €4.3 billion in fraudulent payments in 2022 and 2 billion in the first half of 2023.
Ireland, home to global financial services and tech giants like IBM, BNY Mellon, Stripe, Mastercard, and Citi, remains a hub for the cybersecurity and artificial intelligence with banks and payment platforms increasingly exploring AI solutions to thwart the efforts of fraudsters around the world and protect their customers.
Why AI is a game-changer in fraud detection
Traditional fraud detection systems rely heavily on manual processes and static rule-based systems. These can be time-consuming and are often plagued by high false-positive rates, which are essentially legitimate transactions that are incorrectly flagged as fraudulent.For instance, if a customer who has never withdrawn over €500 suddenly attempts to withdraw more than €1,000, the transaction could be blocked despite being legitimate. Such false positives can lead to unnecessary delays and negatively hit customer satisfaction.
Traditional systems also struggle to quicky process the growing volume of data generated each minute.
In contrast, AI-powered fraud detection systems can analyse trove of transactional data to recognise complex fraud patterns that traditional methods might miss. They can identify unusual activity with far better accuracy. Moreover, AI algorithms continue to learn and adapt once trained, enhancing their ability to detect emerging fraud tactics.
How global giants are driving innovation in AI fraud detection
Tech giant IBM, which has a significant presence in Dublin, helps financial institutions stay ahead of evolving threats by improving fraud detection and reducing false positives. Its real-time payment fraud monitoring solution, IBM Safer Payments, prevents fraud in cashless payment systems like credit cards by analysing transaction streams and intercepting high-risk transactions before they are completed.Irish-founded payment platform Stripe, headquartered in Dublin for European operations, deploys cutting-edge AI tech to shield businesses against online credit card fraud. Powered by adaptive machine learning, its fraud prevention solution – Stripe Radar – absorbs data from the entire financial stack, including payment details and information from giants such as Visa, Mastercard, and American Express.
Global payment giant Mastercard, which operates a major technology hub in Dublin, is also at the forefront of the battle against fraud. It uses AI to enhance the accuracy of real-time transaction approvals and minimise false positives.
Its risk monitoring solution, Decision Intelligence, assigns a score to each transaction, helping banks safely approve nearly 150 billion transactions a year. The AI-powered tech determines the risk score in less than 50 milliseconds by scanning a trillion data points, such as a cardholder’s name, address, purchase history – predicting whether a transaction is genuine.
“With generative AI we are transforming the speed and accuracy of our anti-fraud solutions, deflecting the efforts of criminals, and protecting banks and their customers. Supercharging our algorithm will improve our ability to anticipate the next potential fraudulent event, instilling trust into every interaction,” said Ajay Bhalla, president of Cyber and Intelligence at Mastercard.
Impact of AI solutions in mitigating fraud
According to Mastercard, AI enhancements have improved fraud detection rates on average by 20% and as high as 300% in some instances. Similarly, US banking giant BNY Mellon has improved fraud detection accuracy by 20% using NVIDIA’s DGX AI systems.
PayPal has enhanced real-time fraud detection by 10% running on NVIDIA GPU-powered inference while lowering server capacity by nearly 8x.
UK-based banking giant HSBC, which checks about 900 million transactions for signs of financial crime each month, announced it reduced false positive cases by 60% and detected two to four times more financial crime than previously, following a tie-up with Google to co-develop an AI system called Dynamic Risk Assessment.
Irish pillar banks investing in AI fraud detection tech
In 2024, Bank of Ireland announced an investment of €50 million in fraud prevention and protection, including €15 million on technology to fight scammers.AIB, one of the largest Irish banks, signed a €65 million deal with IBM in 2021 to bolster its cyber resilience and fraud detection capabilities.
Permanent TSB has partnered with Expleo to integrate an anti-fraud software feature to its banking app. The AI-powered feature has been designed to protect users from smishing or SMS phishing by sending alerts when a fraudulent text is detected.
These investments assume significance as Ireland has the highest rates of phishing, a cyberattack method that targets individuals through emails, text messages and other forms of communication to steal personal or financial information.
According to Worldwide Independent Network of Market Research findings, released in 2024, 64% of Irish adults experienced phishing, far higher than the global average of 34%. The survey also found out that around 22% of Irish adults surveyed had their credit card or bank account hacked, while 13% experienced email account breaches.
AI in fraud detection: Implications and challenges
While AI holds immense potential in detecting fraud in the banking and payments space, it has also enabled fraudsters to elevate their tactics.Cybersecurity firms face an ongoing challenge to stay ahead of fraudsters who now exploit AI to generate deepfake videos, synthesised audio and convincing phishing emails to commit fraud.
Companies must also navigate compliance obligations and ethical implications of using AI in fraud management. Misuse or biases within AI systems can cause unintended harm, making responsible AI deployment critical.
In Europe, companies must comply with the European AI Act and the General Data Protection Regulation (GDPR). While the AI act sets out a clear set of risk-based rules for AI developers and deployers, GDPR ensures that organisations handle personal data responsibly while safeguarding individuals’ privacy rights.
FAQs: AI in fraud detection
Are banks in Ireland using AI for fraud detection?Many banks and financial institutions in Ireland use AI to combat fraud.
How does AI improve fraud detection compared to traditional methods?
AI reduces false positives and detects complex fraud patterns in real-time by continuously learning from large volumes of data.
Which industries in Ireland are leading AI adoption for fraud prevention?
The banking and payments sectors, along with fintech companies, are leading AI adoption for fraud prevention in Ireland
What impact has AI had on fraud detection success rates?
AI has boosted fraud detection rates by 20% to 300% in some cases, while significantly reducing false positives.