SWIFT CEO Pushes Information Sharing, Improved Security

Threat Intelligence Sharing Moves Center Stage for SWIFT

SWIFT CEO Gottfried Leibbrandt today revealed a five-point plan designed to harden SWIFT transactions following the $81 million theft via the Bangladesh central bank. Speaking at the 14th annual European Financial Services Conference in Brussels, Leibbrandt focused mostly most on the need for the global financial community to improve its threat intelligence sharing.

This comes as no surprise. A BAE Systems report on the incident indicates that it discovered indications of more than just the Bangladesh attack on SWIFT member banks. It mentioned a bank in Vietnam. It did not name the bank; however it did indicate that the attack took place several months before the attack against the Bangladesh bank. 

Soon after, Hanoi-based Tien Phong Bank (TPBank) released a statement saying that it had interrupted the attempted theft of approximately $1.1 million via fraudulent SWIFT messages.

If TPBank did not report the attack to its own State Bank of Vietnam (after all, it was a failed attack) then it may not have reported it to SWIFT. This leaves an open and purely hypothetical conjecture: if TPBank had shared the information with SWIFT, and if SWIFT had a sharing mechanism with the rest of the member community, could the Bangladesh theft have been prevented?

Whether this conjecture is valid or not, SWIFT is now moving to address the issue. Point one of the five point plan is to ‘drastically improve information sharing’. “We will demand more information of our customers,” said Liebbrandt, “and share that back with the community. The ambition is to do on an international scale what banks in several countries are already doing domestically. We will do it in a confidential way that uses the data while protecting the identity of the institution and customers.”

Not one of the other four points was given more than a single sentence. They are, to “harden security requirements for customer-managed software”; to “develop security audit frameworks for customers”; to “support banks’ increased use of payment pattern controls to identify suspicious behavior”; and to “introduce certification requirements for third party providers”.

None of these intentions affects the SWIFT network itself. Indeed Liebbrandt stressed again (he has done so before) “SWIFT, our network, software and our core messaging services have not been compromised.”

“In Bangladesh and the other cases, the thieves compromised the IT environment and worked their way to the bank systems where the SWIFT instructions are generated and the confirmations received,” Liebbrandt said. “And while we (and other providers) give tools and software to our customers, our customers run these in their own environment and need to keep them secure. We cannot secure our customers’ environments and cannot assume responsibility for that.” 

Nevertheless, this five-point plan demonstrates that SWIFT has learned, albeit the hard way, that any network is only as strong as its weakest link. SWIFT is now taking steps to strengthen those weak links.

Related: Learn More at the 2016 CISO Forum, June 1-2, 2016