A guest editorial by Tamas Kadar, co-founder and chief executive officer of SEON
Real-time payments have transformed the financial landscape, bringing speed and convenience that consumers and businesses expect.
But as payments get faster, so does fraud. Accelerated by AI-driven, large-scale criminal tactics, financial institutions are facing immense pressure to detect and prevent fraud in an instant. Yet, 38% are failing to shift away from traditional batch-based monitoring in favour of real-time transaction monitoring, simply relying on outdated methods such as device checks or two-factor authentication.
Stopping fraud in real time requires more than just faster detection; it demands a fundamental shift in how financial institutions approach risk, security and resilience. Fraud has always been a game of cat and mouse, but now it feels more like a high-speed chase.
The question is: how can businesses keep up? To truly understand the challenge, we need to pull back the covers and examine the evolving landscape of fraud.
Pulling back the covers of fraud
Fraud is a systemic risk that significantly impacts customer trust, operational efficiency and long-term growth. While most organisations focus on the immediate costs of chargebacks or stolen funds, the ripple effects – from compliance breaches to reputational damage – can be far more damaging. Businesses may be losing up to 5% of their revenue when factoring in operational inefficiencies, compliance costs and customer churn. The evolution of fraud tactics requires businesses to reevaluate their approach to measuring the true cost of fraud, as traditional metrics often fall short.
Despite 85% of companies increasing their fraud budgets, over half (56%) still don’t believe fraud is outpacing revenue growth – a sign that many may be underestimating its true financial impact. This suggests a significant blind spot in how fraud costs are calculated, particularly given that 43% of businesses report fraud growing faster than revenue. However, 20% of organisations specifically track direct revenue loss as a key measure of their total fraud and risk management costs. This disparity underscores the problem of many businesses prioritising short-term financial metrics over operational inefficiencies and brand reputation, underestimating the full impact of fraud on their bottom line.
This underestimation is particularly dangerous as fraudsters are rapidly adapting and exploiting new vulnerabilities. Focus is shifting to emerging financial technologies, with digital wallets (48%) and cryptocurrency (30%) among the most targeted fraud frontiers in 2025, particularly as more young people invest in crypto assets. While digital payment solutions, such as WERO, offer benefits from lower transaction costs to deeper insights into consumer behaviour, they also introduce new vulnerabilities that fraudsters are quick to exploit. Additionally, the industry must address the vulnerabilities of these new payment rails.
Meanwhile, regulatory mandates exacerbate the challenge financial institutions find themselves in. The EU’s Instant Payments Regulation (IPR), for example, requires all EU banks to have received instant payments as of January 2025, with a deadline to support sending them by October. This transition brings significant operational and security challenges, including payee verification, bulk payment processing and liquidity management. With these changes requiring significant infrastructure investment, it’s no surprise that two-thirds of European banks report challenges meeting the 2025 EU regulations for real-time transactions.
Failure to comply, particularly ahead of the IPR October deadline, could further erode consumer trust. For institutions still relying on batch-based fraud monitoring, the challenge is even greater – traditional fraud detection methods are too slow for real-time transactions, creating significant vulnerabilities. Put simply, as fraud tactics evolve and regulations increase, so too must fraud prevention strategies. Financial institutions should see this as a chance to innovate. Legacy detection methods relying on point solutions and reactive measures are no longer enough.
A smarter approach to fraud prevention
As fraudsters become more sophisticated, businesses must move beyond fragmented fraud prevention strategies and adopt a more integrated, AI-powered approach. Real-time payments are only growing in popularity, and financial institutions that fail to modernise their fraud defences will not only be left vulnerable but also left behind by the competition. Thankfully, 62% of fraud teams identify real-time transaction monitoring as their most critical investment for 2025.
With that investment comes regulatory alignment, just one key factor in this high-speed race against fraud. But it must be seen as more than just a mere tick-box exercise. Initiatives like the Centre for Finance, Innovation and Technology (CFIT) Digital Company ID concept offer a proactive approach to securing the digital ecosystem within the UK, reducing fraud risks by verifying business identities more effectively. This aligns with the broader push for data standardisation, exemplified by ISO 20022, which enables secure and efficient financial messaging.
Therefore, a holistic, forward-thinking, technology-driven approach that blends AI-driven automation with human expertise is critical. AI-powered fraud detection enhances speed and accuracy, instantly analysing vast amounts of transactional data to spot anomalies. However, human analysts remain essential for interpreting complex cases and adapting fraud prevention strategies to evolving threats. That’s why we’re seeing a staggering 88% of fraud teams increasing headcount as they grow over the next 12 months.
While large enterprises are expanding their fraud teams to better mitigate the risk – with 59% hiring three or more specialists – smaller institutions can’t keep pace without the right solutions, processes and personnel in place. Even within larger organisations, fraud and AML teams often operate in silos, relying on disparate tools and datasets. This lack of cohesion makes it difficult to detect cross-channel fraud patterns and respond to threats effectively.
To truly ‘keep up’, financial institutions must recognise that real-time data is critical for effective fraud prevention. Without centralised, real-time data systems, they cannot achieve the level of visibility needed to combat fraud efficiently. Access to this intelligence through an integrated platform that serves multiple teams across the business is key. However, this only works if those additional people have the right skills. That’s why 76% of organisations cite the most desirable skills going forward are related to advanced data analytics, AI and machine learning.
Of course, real-time data and AI are crucial components, but not the only ones, in constructing adaptive fraud defences. Fraud prevention must also extend beyond transactions. Fraudsters exploit vulnerabilities at every stage of the customer journey, whether that’s at the onboarding or payment processing stage. As such, financial institutions must invest in real-time detection and decisioning across all touchpoints, making sure they can proactively identify and mitigate threats before they cause harm.
Fighting for the future of real-time payments
The impact of fraud extends far beyond direct financial losses. It creates ripple effects that disrupt operations, erode customer trust and drive-up compliance costs. Financial institutions can no longer afford to rely on outdated or siloed fraud prevention strategies.
However, while 85% of organisations have increased their fraud prevention budgets, investments must be strategic, proactive and built for the future. Real-time transactions are here to stay, and fraudsters are adapting just as quickly. After all, it’s about whether your defences can keep up with the changing times.
The time to act is now. Financial institutions that take decisive action today will be the ones that stay ahead, protecting both their bottom line and their customers in an increasingly high-speed threat landscape.
Image: Jakub Żerdzicki on Unsplash
Guest Editorial
This article was produced specially for Fintech Intel by an expert guest contributor.
