As card transaction volume grows, so does the level of fraud associated with those transactions.

In fact, growth in fraud is outpacing the growth in electronic transaction volume, according to new research from The Nilson Report. The publication said in a press release that global card fraud losses grew by 20.6% to total $21.84 billion last year, while transaction volume grew by 7.3%. 

With statistics as grim as this, a smart approach to fraud detection is on many payment facilitators’ minds.

Stripe made headlines in this area last week when it announced its new fraud prevention tool, Stripe Radar. Radar uses a type of artificial intelligence called machine learning, which enables the tool to continuously learn from new data without intervention from humans.

The company has integrated the tool into its system, so merchants using the Stripe platform can use the service immediately.

“Because Radar is integrated with users’ Stripe accounts, it works for every user from day one, without any setup or integration work required,” Stripe’s head of machine learning, Michael Manapat, said. “By contrast, legacy fraud tools require lots of time and effort to set up hundreds (or thousands) of rules, and then large in-house teams to manually review charges. And more modern third-party tools that use machine learning still require weeks or months for each business to train the models.”

While the service works “out of the box” for merchants who use Stripe’s API, it is also customizable. Merchants are able to set custom rules and review flagged transactions manually from their dashboards.

Stripe Radar uses aggregate data from the businesses across Stripe’s network to inform its fraud monitoring tool. This gives the system a significant amount of information the company says helps improve its accuracy when identifying likely fraud. While a particular credit card may be new to a certain user, Stripe will have seen the average card six times before, Manapat said.

The company said in a press release that Stripe Radar was able to block more than $40 million of attempted fraud for Watsi, a non-profit that helps fund medical treatments for people around the world, over the course of a two-month beta test.

Stripe Radar is currently offered to Stripe users at no additional cost.