Citi’s AI-powered tool for catching suspicious payments and thwarting cyber crime is opening up to customers across the world

US Cyber Command

Citigroup is going global with an artificial-intelligence backed tool that helps giant corporations and governments flag aberrant or suspicious payments.
The bank started piloting its Citi Payment Outlier Detection tool in late 2017 to better protect the $4 trillion in payments the bank’s thousands of corporate and public-sector customers send and receive every day.
That fast-paced business is increasingly under threat from cyber crime.
After a successful trial with 20 large clients, including Xerox and Swiss industrial conglomerate Tetra Laval International, Citi is launching the feature in 90 countries.
Click here for more BI Prime stories

After a successful trial run that struck a chord with clients, Citigroup is going worldwide with an artificial-intelligence backed tool that helps giant corporations and governments flag aberrant or suspicious payments and contain the damage from cyber attacks.

To safeguard its customers and streamline payments processing, the bank in late 2017 began piloting a tool called Citi Payment Outlier Detection to help identify unusual money transfers — say, a $5 million transaction sent at an odd hour of the night or to an unexpected country.

The bank initially launched the feature with select customers within its Treasury and Trade Solutions division, which processes $4 trillion in payments each day for thousands of companies and public institutions across the world, and now has more than 20 clients on board, including Xerox and Swiss industrial conglomerate Tetra Laval International.

After more than a year of tweaking and enhancing the tool and receiving encouraging reviews from its pilot group, Citi’s treasury division has now launched the feature in 90 countries and opened it up to all its clients, the company announced Wednesday.

Alert tools like Citi’s aren’t new in the industry and are only increasing in importance as cyber attacks roil corporations. A record 78% of organizations fell victim to payment fraud in 2017, according to a survey from the Association for Financial Professionals.

But most solutions detect payment outliers based on general sets of rules, which can’t account for idiosyncrasies of individual companies with rapidly evolving and expanding business needs, according to Manish Kohli, global head of payments and receivables at Citi.

Aberrant payments can have a number of explanations, some innocent — an employee made an error and fat-fingered a transaction — and some nefarious, such as an unfolding cyber attack.

A rules-based approach can result in false positives that gum up a process that moves at a rapid pace and is only getting faster amid the adoption of digital and automated-payment technology. It can also numb the customer to the alerts, leading to them letting their guard down.

"Companies get into new business lines, they start interacting with new suppliers, they start centralizing operations. So many changes happen that, in the context of payments, having a rules-based system doesn’t work well," Kohli told Business Insider.

What’s unique about Citi’s solution, the company says, is that it’s tailored to the behaviors of particular companies and automatically adjusts its controls, thanks to the machine-learning technology that underpins the service. True anomalies can be flagged for review with precision prior to clearing a payment, and false-positives are minimized.

Feedback from pilot companies convinced Kohli that they’d found a sweet spot with clients that was worth further investment and expansion.

In December, Citi partnered with Feedzai, an AI-powered fraud-detection start-up, in part to further bolster the capabilities of the payment outlier detection feature, which was initially created in Citi’s Dublin innovation lab.

"The feedback was very varied, but it was resounding and reassuring enough that we’d hit on something very valuable," Kohli said. "We touched on a pain point for clients that was interesting enough for us to invest in — not just making it more global, but enhancing its functionality."

The payment-detection tool integrates into Citi’s existing digital payments platform but comes at an additional cost, which will vary depending on the size and complexity of individual organizations.

NOW WATCH: MacKenzie Bezos pledged to donate more than half of her life’s fortune. Here’s how she went from one of Amazon’s first employees to an award-winning novelist.