Artificial intelligence models are only as effective as the data they’re trained on, which is one reason why Stripe believes its AI-driven payments platform can better detect fraud.
At an event, the company said its Payments Foundation Model has been trained on billions of transactions that flow through its systems, which makes the AI model more attuned to the nuanced aspects of each transaction.
One example is card testing fraud, where criminals run small transactions to check if stolen card details are still active. Stripe said that while its previous AI tools had some success in blocking this kind of fraud, the new model could reduce card testing by 64% almost immediately—thanks to expanded access to the company’s transaction data.
Following in the Footsteps
Stripe is following in the footsteps of some of the world’s largest financial players, who are doubling down on their AI initiatives.
Both Mastercard and Visa have launched new platforms designed to capture the potential of agentic AI. Mastercard’s Agent Pay and Visa’s Intelligent Commerce platforms are built to handle all the aspects of a transaction autonomously—from picking out items to the final purchase.
In the crypto space, Coinbase has unveiled its x402 payments mechanism that leverages an existing HTTP protocol to enable both humans and AI agents to conduct stablecoin transactions during web interactions.
Replacing the Coach
As hot as AI is, stablecoins have also been making headlines in recent months. After PayPal launched its stablecoin two years ago, it seemed natural that Stripe would follow suit with one of its own. This launch became inevitable after the company’s billion-dollar acquisition of stablecoin company Bridge.
However, Stripe has broader ambitions in the stablecoin market. The fintech’s leadership has indicated plans to bring stablecoin-backed, multicurrency cards for businesses. The goal is to give businesses in different countries the ability to operate using the same currency.
Additionally, Stripe is planning to roll out a range of new offerings, including everything from tax help to instant payment integration. However, it’s unclear whether this bevy of solutions will help the company move forward.
“Stripe is persistent, if nothing else, as it relentlessly chases global omnichannel merchants,” said Don Apgar, Director of Merchant Payments at Javelin Strategy & Research. “The press releases, product stack and required features/functions are all there, the only thing missing is the large enterprise merchants.”
“Like the sports team that continually replaces the head coach, at some point you have to wonder what the real issue is,” he said. “However, this fraud model could be a game-changer if it truly delivers the results that Stripe claims.”
The post Stripe’s AI Model Touted to Be More Effective Against Fraud appeared first on PaymentsJournal.
