AI-powered fraud prevention: tackling identity theft, payment fraud, and scams in real-time

Cleber Martins, Head of Payments Intelligence & Risk Solutions at ACI Worldwide, explains how AI and machine learning enhance fraud detection by analysing real-time data and enabling adaptive, collaborative intelligence.

Given the increasing sophistication of cybercrime and the significant financial losses due to fraud, how can AI and machine learning solutions enhance fraud detection and prevention for financial institutions, especially in identifying and mitigating emerging threats like identity theft, payment fraud, and scams across various channels and payment methods?

Identity theft, scams, and payment fraud are closely linked, with criminals often targeting identities to create accounts for moving and controlling funds. They exploit new payment methods to access cash more quickly, bypassing traditional steps like selling products on the black market, which was common in ecommerce fraud.

This issue is escalating, as criminals find new ways to access and move funds. Over the past 20 years, banks have implemented robust security measures to ensure only account owners can initiate transactions, keeping systems secure. However, scams - especially Authorised Push Payment (APP) fraud - have changed the landscape. In APP fraud, criminals manipulate customers into making payments themselves, bypassing the bank’s security controls, as the victim believes the payment is legitimate, making fraud detection increasingly difficult.

This is where artificial intelligence (AI) becomes crucial. Fraud detection requires analysing massive amounts of data in real-time - something beyond human capability. Criminals often use social engineering tactics, such as creating a sense of urgency or exploiting inherited trust by posing as someone the victim knows. AI can detect subtle behavioural signals, like a rushed transaction, that indicate fraud, allowing banks to intervene in real-time. AI advancements are necessary to identify such patterns quickly, a task that would be difficult for a human observer to catch.

Without advanced AI, banks face challenges such as blocking transactions based on recipient or amount, which can frustrate legitimate customers. As the demand for faster and more reliable payment methods grows, outdated fraud detection systems risk alienating customers. Meanwhile, criminals also use AI to scale their operations, targeting vulnerable individuals and creating deepfake voices to impersonate loved ones, making scams harder to detect.

This has created a technological arms race, where criminals can deploy advanced AI tools, and banks must invest in real-time, adaptive AI solutions. Relying solely on historical data is no longer enough. The next generation of AI will enable real-time, collective learning, allowing financial institutions to share insights and adapt to emerging threats faster.

With the rapid rise of immediate payments and scams, the financial and legal implications are high. The future of fraud prevention lies in real-time, adaptive AI systems that protect customers and keep pace with evolving threats.

Considering the complexities of global payment ecosystems and the need for real-time fraud detection, what strategies and solutions can help banks and payment processors effectively analyse large volumes of transactional and behavioural data, detect anomalies, and prevent fraud in cross-border payments, while complying with diverse regulatory requirements and ensuring a seamless customer experience?

This question is crucial because it highlights one of the biggest challenges we’ll face as we move into this new era of collaboration. As I mentioned earlier, the key is leveraging technology that allows banks to learn together, without compromising their customers, data, or internal strategies. Rather than centralising all the data in one place, the focus should be on bringing the learning to where the problem exists, instead of trying to collect and analyse everything in a single location. Given the complexity of modern financial fraud, a decentralised, real-time approach is essential.

From a technology perspective, many of these new payment methods are built on the ISO 20022 protocol- a highly data-centric payment transportation method. Unlike traditional payment systems, ISO 20022 allows transactions to carry far more information than just the sender, receiver, and transaction amount. This protocol can embed AI-driven signals - intelligence collected at the point of transaction initiation - into the payment itself in a format that is not human-readable, almost like a digital DNA.

This advancement allows financial institutions to exchange intelligence in real-time without compromising customer privacy or revealing their internal fraud prevention strategies. It fosters an ecosystem where banks can collectively learn from each other, strengthening fraud detection across the board.

The future of payments and fraud prevention lies in real-time intelligence exchange. AI systems must be able to:

1. Transport and consume real-time fraud signals embedded in payments.
2. Correlate these signals with existing data from the initiating bank, central payment infrastructures, and the receiving institution.

Central payment infrastructures, which act as intermediaries for new payment methods, play a critical role in this ecosystem. They have a unique, independent view of transactions flowing through the system. By combining intelligence from:

• The initiating bank (which sees customer behaviour and account activity),
• The central infrastructure (which monitors transaction flow at a broader scale), and
• The receiving institution (which has insight into account ownership and potential fraudulent activity on the other end),

these three entities can work together to make precise, real-time decisions about potential fraud.

One of the biggest challenges in combating scams, especially APP fraud, is that criminals have full visibility of the entire fraud scheme, while financial institutions only see fragments of the process.

Criminals understand:

• Who they are attacking and how to manipulate them.
• Which bank products and processes they need to exploit to bypass security controls.
• How the transaction will flow and what weaknesses exist along the way.
• How to retrieve and launder the stolen funds, whether through cash withdrawals, cryptocurrency, or other means.

Right now, financial institutions and payment networks are playing catch-up because they only see pieces of the puzzle, whereas criminals see the entire picture. However, real-time intelligence sharing through embedded fraud signals can change that.

By combining intelligence from multiple sources, banks and payment networks can achieve greater visibility than criminals, allowing them to detect and block fraudulent transactions faster than ever before. When fraudsters encounter a system that predictively blocks their schemes in real time, they’re forced to abandon those tactics and move elsewhere.

This is the future of fraud prevention - one where AI-driven, real-time collaboration makes financial ecosystems more resilient than ever before.

Given that various reports estimate the global cost of fraud to be in the trillions of dollars annually, impacting GDP, household finances, and even fuelling other crimes, how does ACI Worldwide help its clients quantify the economic impact of fraud reduction through their services? Additionally, what innovative approaches can be used to further minimise the broader societal and economic consequences of fraud on a global scale?

Fraud is a major global issue impacting trust, financial institutions, and economies. ACI Worldwide, a leader in payments for over 50 years, has driven real-time, account-to-account payments that are faster, more cost-effective, and enhance financial inclusion, positively impacting GDP by putting money back into the economy. Many governments support next-generation payment systems for these benefits.

To help the industry understand fraud reduction’s economic benefits, ACI collaborates with specialists to produce research on global trends in real-time payments. These reports also highlight fraud evolution and criminals adapting to exploit new methods, with ACI's Scamscope report tracking scams' rising impact on the financial ecosystem.

ACI works with financial institutions globally, leveraging cross-market intelligence to help partners stay ahead of emerging fraud tactics. Collaboration and AI-driven intelligence sharing are key to combating fraud. ACI empowers real-time fraud detection by fostering seamless collaboration between institutions, allowing them to stop fraud before it spreads. AI-powered models integrate localised fraud prevention with global intelligence, continuously learning and adapting to ensure swift, effective responses.

By combining expertise, AI-driven intelligence, and global collaboration, ACI Worldwide leads the fight against fraud, helping financial institutions stay ahead of emerging threats and strengthening the financial system.

This interview was originally published in the Trends Report: The New Payments Era. Download the report and start building your next-generation payments strategy today.

About the interviewee

Cleber Martins joined ACI Worldwide in 2001 and has 20 years of experience in implementing industry-leading enterprise fraud prevention solutions and anti-money laundering strategies. Cleber's enthusiasm for driving innovation in fraud prevention stems from the pride he takes in protecting both his banking customers and the people and communities they serve.

Cleber's key areas of expertise include helping payments leaders to develop multi-faceted fraud prevention strategies that combat modern threats, as well as creating actionable intelligence in payments data, and the evolution of fraud prevention into a Customer Experience differentiator for organisations. 

About ACI Worldwide

ACI Worldwide, an original innovator in global payments technology, delivers transformative software solutions that power intelligent payments orchestration in real-time so banks, billers, and merchants can drive growth, while continuously modernising their payment infrastructures, simply and securely. With nearly 50 years of trusted payments expertise, we combine our global footprint with a local presence to offer enhanced payment experiences to stay ahead of constantly changing payment challenges and opportunities.