How anonymized and aggregated transaction data powers new AI models
Imagine if you could leverage the business intelligence of one of the largest transaction datasets in the world. Trained on Mastercard’s anonymized and aggregated global transaction data,* Mastercard’s new market-ready artificial intelligence (AI) models powered by Brighterion AI are revolutionizing transaction fraud prevention.
Financial institutions (FIs) face fraud and credit risks every day and must be vigilant to avoid losing millions of dollars. Yet according to Payments Journal, many transactions that acquirers identify as fraudulent are actually good transactions. This critical factor hurts merchants’ bottom lines in lost revenue and reputation from consumers that see merchants and issuing banks as the problem.
Now these problems can be managed quickly and accurately with out-of-the-box AI solutions that are ready to deploy globally in as little as 30 days. The risk of more time-consuming custom model building has been removed and the results speak for themselves.
With Mastercard’s global network of 210 countries and territories, the breadth of transaction data is vast. Using transaction data for financial data analytics while respecting customer privacy has become a core competency for Mastercard. The financial data does not contain identifying factors; the personal account information, geography, merchants and other identifiers have been stripped out. FIs can benefit from models trained on this data, which can learn patterns and anomalies and instantly score transactions to deliver intelligence in real time.
Turning AI expertise and experience with transaction data into the future of AI for fraud and risk
Brighterion AI has proven its accuracy in preventing transaction fraud on the Mastercard network and is the backbone of Mastercard’s award-winning Decision Management Platform.
With over 20 years of experience in AI innovation, the team strategized a way to develop faster-to-implement models. Building on Mastercard’s 55+ years of payments experience, these out-of-the-box models are pre-trained with global network intelligence derived from anonymized and aggregated transaction data.
Pre-trained AI models don’t have to refer to network data during transactions or to score events. This enables the cloud-native platform to deliver a remarkably low latency of 100-120ms and, when deployed on-premises, a speed of less than 10ms.
Global transaction data helps acquirers detect fraud faster
These more timely, broadly trained models are beneficial to acquirers, payment processors and payment facilitators.
Global transaction data brings knowledge from the broader market to deliver fewer false positives and higher fraud detection rates. Using Brighterion AI’s market-ready solution, one acquirer saw an increase in fraud detection of two to three times and an increase in approval rates of 7.4 percent.
The models provide a 360-degree view of acquirers’ merchant ecosystems, covering risks such as fraud and collusion. Acquirers can monitor merchants for viability and risk while reducing friction for trusted users.
How does training AI on Mastercard transaction data differ from using an FI’s own historical data?
Building traditional/custom AI models
Traditional, custom AI models require a longer turnaround time. Some FIs struggle to extract the right financial data for model building and training or may not have the historical data that developers need.
FIs are burdened with extracting huge datasets to meet this need, ensuring correct labeling and effective data transfer. The required datasets may be in the hundreds, requiring hours of work. These are used to train the new model to identify anomalies relevant to the business’s specific challenges.
Initializing self-learning, market-ready AI models
Market-ready models are built with a variety of advanced AI and ML technologies. These off-the-shelf solutions move beyond the intelligence found in an FI’s historical data; they are trained with the business intelligence derived from more than 150 billion transactions a year processed by Mastercard. By using this more robust dataset, the model has expanded knowledge.
The key advantage of market-ready AI is that it saves FIs time and resources as the model is already built and trained, reflecting high accuracy rates. The model is ready to deploy with a small sampling of the FI’s transaction data. During this period, the customizable API interface is fine-tuned to the customer’s needs. Scores are then returned to the FI and the model is ready for full deployment.
Market models for acquirers are brand-agnostic. FIs benefit from the scope of business intelligence, not the network brand involved. Market-ready and custom models are self-learning and trained faster with more comprehensive datasets. They continue to learn and evolve with each new transaction or output, with oversight by Mastercard computer scientists.
Save money by blocking fraud sooner
Acquirers save money by deploying transaction data solutions early in the pre-authorization stage rather than at the transaction stage. By blocking fraud before it hits issuer or network solutions, acquirers are:
- Clarifying where the fraud liability lies, whether the issuer, network or acquirer
- Saving money that would be lost to transaction costs, merchant fees and chargebacks
- Increasing accuracy of prediction rates
- Decreasing fraud rates to protect accuracy ratings
- Preventing the most undetectable fraud that would have been cleared at other stages of the transaction process
- Reporting a 15 to 30 percent improvement in detection rates
A place for custom AI
Mastercard has perfected the process of building custom models using the Brighterion AI platform, preparing customers for deployment in a couple of months with our AI Express implementation process.
Custom-built AI models don’t have to be complicated and are still the right fit for specific or unique business challenges where innovation and experimentation are needed.
Do market-ready models spell the end of custom AI? No, but for certain use cases, market-ready AI can save FIs time and money.
Conclusion
Market-ready AI models trained on Mastercard’s anonymized and aggregated global transaction data are another innovation that continues to improve results and save customers time and money. These models offer the advantages of globally accumulated business intelligence to deliver low latency and high accuracy. Mastercard’s market-ready models are brand-agnostic and ready to operate on most FIs’ systems.
Learn how market-ready solutions built with Brighterion AI are production-ready out of the box to deliver superior financial analytics and immediate return on your investment.
*All Mastercard transaction data has been aggregated and anonymized when used to build AI and ML models on the Brighterion AI platform.
