Introducing the Airwallex Payment Fraud Engine: Stop fraud without killing conversion

As the digital economy expands, protecting online payments has become more complex, and more critical, than ever. Card-not-present transactions, subscription models, digital goods, and cross-border eCommerce have made it easier for fraudsters to target online payments. At the same time, card scheme programs like Visa VAMP and Mastercard monitoring are tightening their thresholds, and customers expect a fast, low-friction checkout every time.

Most legacy fraud tools force you into a painful trade-off. If you block aggressively, you may end up losing legitimate customers and revenue through false declines. And if you relax controls, you may absorb greater fraud losses, chargebacks, and scheme risk.

Today, we’re introducing the Airwallex Payment Fraud Engine, built to eliminate that trade-off. The Airwallex Payment Fraud Engine combines advanced AI and machine learning, real-time rules, and deep integration across the payment stack to:

• Stop fraud and coordinated attacks in real time before they hit your P&L.

• Minimise false declines by triggering step-up authentication (e.g. one-time passcodes, banking app challenges) only when needed.

• Protect your standing with card networks across markets and card ranges, and stay within compliance programs.

In 2025 alone, the engine prevented 45% more fraud losses year-on-year while improving payment success and reducing false declines.

A unified risk and revenue protection stack

The Payment Fraud Engine doesn’t operate in isolation: it’s part of a unified, AI-powered payments stack designed to protect revenue end-to-end. Instead of being a standalone fraud tool, it sits at the heart of a tightly integrated system:

• Payment Fraud Engine: Screens and scores every transaction in real time and returns an allow/challenge/block decision. Post-payment outcomes are continuously fed back to improve performance over time.

• Optimize 360: Chooses the best payment path, including retries, security checks, and tokenization strategies to maximise approval rates.

• Airwallex Sentinel: Detects coordinated attacks and deploys targeted strategies in real time, with clear visibility into fraud prevented.

• AI Dispute Automation: Recommends whether to accept or challenge disputes and automates evidence submission to recover revenue.

Because Airwallex controls acquiring, fraud, authentication, routing, tokenization, and disputes, all components work together: fraud decisions feed directly into routing and authentication; attack detection informs real-time strategies; and dispute outcomes continuously improve model accuracy.

The result: fraud prevention, optimisation, and dispute management working as a single system instead of disconnected tools.

Inside the Airwallex Payment Fraud Engine

At a high level, the engine evaluates every payment in real time and determines the optimal action without adding unnecessary friction.

Data foundation and risk signals

High-quality data underpins every decision. Airwallex consolidates a rich set of signals into a unified risk data layer, including payment and card details, device fingerprinting and behavioural signals, customer profiles, and network-level data.

This standardisation gives the engine a consistent, high-confidence view of each transaction while improving model accuracy, reducing false declines, and helping maintain healthy scheme performance.

Real-time decisioning pipeline

When a payment reaches the fraud engine, it flows through a real-time pipeline designed to assess risk without slowing checkout.

First, the engine maps each transaction to relevant risk profiles, such as digital goods, travel bookings, or recurring payments, so decisions are tailored to the context rather than applied uniformly.

It then analyses and evaluates thousands of signals, from long-term behavioural patterns to real-time velocity and network-level activity, to generate a precise risk score. That score feeds into a layered rule system, where safeguards for known attack patterns are combined with segment-specific strategies, scheme-aligned controls, and merchant-defined rules. This ensures decisions aren't only accurate, but also compliant and adaptable to different risk appetites.

Finally, the fraud decision is orchestrated alongside payment optimisation. This determines whether to trigger additional authentication, which acquiring route to use, and whether a retry should be attempted. Together, this approach treats fraud as a core part of payment optimisation, helping improve conversion rather than simply blocking risk.

Merchant controls and transparency

Automation is most effective when paired with visibility and control. Airwallex provides merchant-facing tools that make it easy to understand and fine-tune performance over time.

Risk Analytics dashboards surface key metrics like fraud rates, dispute ratios, and approval trends, often benchmarked against similar businesses. Watchlists allow you to explicitly allow or block specific attributes, while structured reason codes translate complex decisions into clear, actionable explanations.

This combination gives you the confidence to rely on automation while still having the control to step in when needed.

2025 performance: What the fraud engine achieved

Across card payment traffic processed by the engine in 2025:

• Higher success rates: +4.24% increase driven by fraud + optimisation

• Lower fraud rates: Reduced by over 60%

• Fewer false declines: Cut by ~50%

• More fraud prevented: +45% year-on-year

These gains were consistent across key segments, including digital goods, travel, and new merchants.

What sets Airwallex apart

Many solutions offer AI scoring, rules, and dashboards. Airwallex goes further by embedding fraud prevention directly into the payment lifecycle.

1. Deep integration with the payment stack

Because Airwallex runs the full-stack, end-to-end, fraud decisions don’t sit in isolation – they directly influence how payments are processed. This means authentication can be dynamically applied only when needed, payments can be routed more intelligently to improve approval rates, and data can be optimised to increase issuer confidence. This leads to higher approval rates and lower fraud, without adding friction or operational complexity.

2. Full lifecycle coverage

Airwallex supports the entire payment journey, from the moment a customer checks out through to dispute resolution. Fraud is stopped before authorisation, decisions are continuously refined based on real transaction outcomes, and if disputes occur, they can be handled with automated tooling that reduces manual effort and improves recovery rates. This end-to-end approach helps protect revenue at every stage (not just at checkout).

3. Proven performance

Internal evaluations and live merchant comparisons show strong results across both fraud prevention and conversion. In multiple scenarios, the Airwallex Payment Fraud Engine has matched or outperformed dedicated vendors, while maintaining low false decline rates and strong overall approval performance. All of this is delivered on a single, unified platform, reducing the need for multiple point solutions.

What this means for your business

Real-time protection

Block fraud and attacks before they impact revenue

Higher approval rates

Let more legitimate customers through with less friction

Lower fraud and dispute costs

Stay within scheme thresholds and recover more revenue

Transparency and control

Understand decisions and fine-tune your strategy

Future-ready AI

Continuously adapt to new fraud patterns without manual effort

Bringing it all together

By unifying fraud prevention, optimisation, and disputes, Airwallex helps you maximise revenue while minimising fraud and operational costs. It protects your standing with card networks and enables you to deliver seamless customer experiences that keep customers coming back.

If you’re looking to replace legacy tools or consolidate multiple vendors, the Airwallex Payment Fraud Engine offers a modern, integrated approach, so you don’t have to choose between growth and protection.