TL;DR:
- Mastercard introduces Decision Intelligence Pro, a proprietary generative AI model.
- The AI enhances fraud detection in real-time for thousands of banks within Mastercard’s network.
- The model utilizes a recurrent neural network and transformer models.
- Data from 125 billion annual transactions drives the AI’s understanding of merchant relationships.
- Instead of text-based prompts, the algorithm relies on cardholders’ merchant visit histories.
- Decision Intelligence Pro improves fraud detection rates by up to 300%.
- Mastercard has invested over $7 billion in cybersecurity and AI technologies in the past five years.
- Visa is also investing in AI, signaling a growing trend in the payments industry.
- The technology has the potential to reduce costs for financial institutions and identify new fraud patterns globally.
Main AI News:
Mastercard, a leading player in the payments industry, has taken a significant leap into the world of generative artificial intelligence with its groundbreaking proprietary model, Decision Intelligence Pro. This innovative AI solution is poised to revolutionize the way thousands of banks within the Mastercard network detect and combat fraudulent transactions, potentially boosting fraud detection rates by up to 300%.
Ajay Bhalla, President of Mastercard’s Cyber and Intelligence Business Unit, revealed exclusively to CNBC the remarkable capabilities of Decision Intelligence Pro. This proprietary recurrent neural network, a cornerstone of generative AI, has been meticulously crafted in-house by Mastercard’s cybersecurity and anti-fraud experts. Bhalla emphasized, “We are leveraging transformer models to harness the full potential of generative AI. Our approach is entirely in-house, supported by a wealth of data sourced from our extensive ecosystem, which is a fundamental aspect of our business.”
While Mastercard occasionally relies on open-source technology when necessary, the majority of this cutting-edge AI technology is developed internally. The proprietary algorithm behind Decision Intelligence Pro is trained on a staggering 125 billion annual transactions within the Mastercard card network. Instead of focusing on textual inputs, the algorithm utilizes a cardholder’s history of merchant visits to assess whether a transaction is consistent with the customer’s typical behavior.
Mastercard’s algorithm then generates pathways within its network, akin to a heat-sensing radar, to provide a score that reflects the transaction’s likelihood of legitimacy. A higher score aligns with the cardholder’s expected behavior, while a lower score deviates from this pattern. Remarkably, this entire process unfolds in a mere 50 milliseconds.
Bhalla highlighted the substantial impact of this new transaction decisioning technology, asserting that it can boost financial institutions’ fraud detection rates by an average of 20%, with some cases experiencing improvements of up to 300%. Mastercard has made substantial investments in the realms of cybersecurity and AI, totaling over $7 billion in the past five years, including strategic acquisitions such as Baffin Bay Networks in March 2023.
Visa, Mastercard’s formidable competitor, has also ventured into the realm of AI, committing $100 million to a venture fund for generative AI startups in October 2023. While it is still in its nascent stages, Mastercard envisions its algorithm could enable banks to save up to 20% by eliminating much of the expenses associated with assessing illegitimate transactions.
Perhaps the most promising aspect of Mastercard’s technology lies in its potential to identify previously unknown fraudulent patterns and trends within the payments ecosystem. Bhalla explained, “The beauty of Mastercard’s ecosystem is that we have a global view of data from all our customers, allowing us to detect fraud and identify patterns on a global scale.”
In an industry where many players are embracing AI, Mastercard’s Decision Intelligence Pro stands out as a formidable innovation with the capacity to reshape the landscape of fraud detection and prevention, ultimately leading to safer and more secure financial transactions.
Conclusion:
Mastercard’s foray into generative AI with Decision Intelligence Pro is a game-changer for the fraud detection market. The model’s impressive ability to boost detection rates by up to 300% and reduce costs for financial institutions by up to 20% positions Mastercard as a leader in the field. Visa’s similar investments indicate a growing industry trend towards AI-driven solutions, making the payments ecosystem safer and more secure.
