Beyond Fraud Detection: The Rise of Intelligent Commerce

IA FORUM MEMBER INSIGHTS: ARTICLE

By Sandip Pandey, Lead Software Engineer (Android), Mobile Systems, Artificial Intelligence & Machine Learning, PAYPAL

How AI is shifting payment systems from processing transactions to understanding customer intent.

One of the things I've always found fascinating about modern payment systems is how much intelligence is packed into a transaction that completes in just a few hundred milliseconds. Behind a simple tap on a phone are countless decisions involving authentication, fraud detection, risk scoring, and payment routing - all working together almost instantly to determine whether a payment should be approved.

For decades, the industry's goal has been straightforward: make payments faster, safer, and more reliable. AI is now expanding that goal.

Most conversations about AI in payments focus on improving fraud detection. While that remains an important area of innovation, I believe the bigger opportunity lies elsewhere. AI is enabling payment platforms to understand the context surrounding a transaction - not just whether it is legitimate, but what the customer is actually trying to accomplish.

A customer booking a flight isn't simply making a payment. They're beginning a journey that may include hotels, transportation, dining, and dozens of related purchases. Viewed individually, these are separate transactions. Viewed together, they tell a much richer story about customer intent.

In my view, this marks the beginning of a new phase in digital payments - one where success is measured not only by how effectively we prevent fraud, but also by how intelligently we help customers complete their commerce journey.

In this article, I share why I believe AI, mobile platforms, and trusted payment infrastructure are converging to make that possible.

Fraud Detection Is Only Part of the Story

Over the past decade, artificial intelligence has transformed fraud detection. Modern payment networks can analyze hundreds of signals - from spending behavior and merchant history to device characteristics and location - in just a few milliseconds before approving a transaction. The result has been smarter risk decisions and stronger protection for consumers and businesses alike.

From an engineering perspective, this is an impressive achievement. Today's payment systems are exceptionally good at answering one important question:

"Can this transaction be trusted?"

That question will always remain fundamental. Trust is the foundation of every payment experience. However, I believe digital commerce is reaching a point where trust alone is no longer enough.

Customers rarely make purchases in isolation. Booking a flight often leads to hotel reservations, transportation, dining, and other travel-related expenses. A homeowner renovating a house may purchase materials, appliances, and services over several weeks. While today's payment systems evaluate these as separate transactions, customers experience them as part of a single journey.

Understanding that journey creates opportunities that extend well beyond fraud prevention. Instead of simply approving or declining payments, AI can help recommend the most appropriate payment method, reduce unnecessary authentication, personalize offers, and recognize legitimate spending patterns with greater confidence.

As AI assistants become more involved in shopping and purchasing decisions, payment platforms will need to understand not only whether a transaction is legitimate, but also whether it makes sense in the broader context of the customer's intent.

In my view, this is the next evolution of payment intelligence - from evaluating individual transactions to understanding the customer journey behind them.

AI Is Teaching Payment Systems to Understand Context

One of the more interesting developments in the payments industry is the emergence of Large Transaction Models (LTMs). In a recent paper, Visa describes how LTMs apply advances in AI to learn patterns from billions of payment events, helping identify relationships that are difficult to capture using traditional rule-based systems.

While the technology itself is impressive, I believe the bigger story isn't the model - it's the shift in thinking that it represents.

For years, payment intelligence has relied on predefined rules and carefully engineered features such as transaction amount, merchant category, device reputation, location, and historical spending behavior. These approaches have been highly effective, particularly in fraud detection, but they still evaluate transactions largely as individual events.

Large Transaction Models introduce a more contextual approach. Instead of looking at a single payment in isolation, they learn from patterns across a customer's broader spending behavior. This enables payment systems to better understand how transactions relate to one another, distinguish normal purchasing journeys from unusual activity, and adapt to changing customer behavior over time.

To me, this represents a natural evolution in payment intelligence. Just as AI transformed search by understanding the meaning behind words instead of matching keywords, payment systems are beginning to move beyond analyzing individual transactions to understanding the broader context in which they occur.

That shift creates opportunities far beyond fraud detection. It lays the foundation for payment experiences that are more personalized, proactive, and capable of supporting the next generation of intelligent commerce.

From Payment Intelligence to Commerce Intelligence

If AI is helping payment systems better understand customer behavior, the next question is: What should they do with that understanding?

In my view, this is where the industry begins to move from payment intelligence to commerce intelligence.

Traditional payment systems are designed to answer a single question: Should this transaction be approved? Commerce Intelligence expands that responsibility by asking another question: How can we help the customer complete their journey more intelligently?

Consider someone planning an international trip. Buying an airline ticket is often followed by hotel reservations, local transportation, dining, and foreign purchases. Instead of treating these as unrelated transactions, an AI-powered platform can recognize them as part of the same journey and make smarter decisions throughout the experience.

That might include recommending the most appropriate payment method, identifying eligible rewards, reducing unnecessary authentication for expected purchases, or flagging activity that genuinely falls outside the customer's normal behavior. The objective is no longer limited to processing payments - it is to deliver a smoother and more personalized commerce experience.

I believe the next architectural evolution in digital payments is what I call the Commerce Intelligence Layer. Rather than replacing today's payment infrastructure, it complements existing payment networks, issuers, and fraud engines by combining customer behavior, device intelligence, historical spending patterns, and AI-driven insights before a payment decision is made.

As illustrated below, this additional intelligence layer helps bridge the gap between customer intent and payment execution. It allows payment systems to move beyond reacting to transactions and begin supporting the customer's broader commerce journey.

To me, this is where AI delivers its greatest value - not by replacing trusted payment infrastructure, but by making every payment interaction more contextual, intelligent, and customer-centric.

Why Mobile Is at the Center of Intelligent Commerce

While much of the discussion around AI in payments focuses on cloud models and payment networks, I believe the real customer experience will be shaped on the device people interact with every day - the smartphone.

Over the years, mobile devices have evolved from simple payment tools into trusted digital companions. Today they combine biometric authentication, secure hardware, encrypted credential storage, location awareness, and increasingly capable on-device AI. No other device understands a customer's context quite as well.

That makes mobile the natural place where intelligent commerce comes to life.

Imagine opening your wallet app before making a purchase. Instead of simply presenting a list of payment cards, it could recommend the best card for that merchant, identify available rewards, recognize that the purchase is part of an upcoming trip, or determine whether additional authentication is actually necessary based on the surrounding context. Rather than adding friction, AI has the potential to remove it.

Another important trend is the rapid advancement of on-device AI. As smartphones become capable of running increasingly sophisticated AI models locally, many decisions can be made without continuously sending sensitive behavioral data to the cloud. This creates an opportunity to deliver highly personalized experiences without compromising customer privacy.

From my perspective, mobile applications will evolve from being interfaces that initiate payments to platforms that actively assist customers throughout their commerce journey. The payment network will continue to authorize transactions, but the mobile experience will become significantly more intelligent, contextual, and proactive.

In my view, this is where AI, mobile technology, and trusted payment infrastructure come together to create the next generation of customer experiences.

Trust Will Be the Competitive Advantage

Imagine asking an AI assistant to plan your next business trip. It books your flight, reserves your hotel, recommends the best payment card, and completes the purchases on your behalf - all within minutes.

Technically, that future isn't difficult to imagine. In many ways, the technology is already emerging.

The bigger question is whether users will trust AI to make those decisions.

For decades, digital payments have earned that trust through security, reliability, and accountability. Customers know their payment credentials are protected, fraudulent transactions can be disputed, and they remain in control of every purchase. As AI becomes more involved in commerce, those same expectations will apply to AI-driven decisions.

In my opinion, organizations should treat governance as part of the product - not simply a compliance requirement. Customers should always understand when AI is acting on their behalf, why a recommendation was made, and when they have the ability to approve, reject, or override a decision.

Privacy is equally important. As more intelligence moves onto mobile devices, there is an opportunity to deliver highly personalized experiences while keeping sensitive customer data closer to the user. This balance between intelligence and privacy will be essential for building long-term trust.

Ultimately, I don't believe the organizations with the most advanced AI models will necessarily lead the next generation of digital commerce. The winners will be those that combine intelligent technology with transparency, responsible governance, and experiences that customers feel comfortable trusting every day.

Closing Thoughts

Artificial intelligence is undoubtedly making payment systems smarter, but I believe its greatest impact will not be in processing transactions faster or detecting fraud more accurately. Its real value lies in helping payment platforms better understand people - their context, their intent, and the journeys behind every purchase.

We've spent decades making digital payments faster, safer, and more reliable. The next decade will be about making them more intelligent.

In my view, intelligent commerce isn't about replacing the trusted payment infrastructure we've built over the years - it's about enhancing it.

Organizations that successfully combine AI, trusted mobile platforms, and responsible governance won't simply process payments more efficiently; they'll redefine how customers experience commerce.

The future of payments won't be defined solely by how quickly money moves. It will be defined by how well technology understands people and helps them achieve what they set out to do. To me, that is the true promise of intelligent commerce.

Author Disclaimer: The views and opinions expressed herein are those of the Author alone and are shared in a personal capacity, in accordance with the Chatham House Rule. They do not reflect the official views or positions of the Author’s employer, organization, or any affiliated entity.