eWEEK content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.
In early February, online thieves were close to stealing nearly a billion dollars from the international banking transfer system run by the Society for Worldwide Interbank Financial Telecommunication (SWIFT) Alliance.
They were stopped, not by the latest in defensive technology nor a well-honed security process, but by a typo. An error in the name of a payee—the non-existent “Shalika Foundation” was misspelled as the “Shalika Fandation”—reportedly raised suspicions at an intermediary bank handling the transactions and delayed the transfers while it sought confirmation. The transactions had also started to raise suspicions at the bank from which the funds were being transferred, the Federal Reserve Bank of New York.
Even so, the thieves made off with approximately $81 million, underscoring that the byzantine checks and countermeasures used by the SWIFT Alliance to keep its network from being abused and its less-than-concrete security guidelines are not enough to stymie a motivated attacker. A second attack against a Vietnamese bank, revealed last week, added an exclamation point.
“What we are seeing from the smaller international or more regional type of banks is a situation where they don’t know what to do to secure their systems against these threats,” Douglas Gourlay, corporate vice president of application-security firm Skyport Systems, told eWEEK. “SWIFT says this is your responsibility, but in their guidance to their members, they have very little that is actionable.”
The incident has cast doubt on the ability of regional central banks to defend against online thieves who demonstrated in-depth knowledge of the international transfer system. The malware suspected to have been used by the attackers did not exhibit any significant sophistication, but showed obvious signs of insider knowledge of how the SWIFT Alliance system worked and the infrastructure of the Bangladeshi bank. In addition, many of the software tools used by attackers are not malware per se, but administrative tools used with malicious intent, making detection of the attacks difficult, security experts said.
Researchers at BAE Systems believe they have identified the malware used to attack the banks. Uploaded to a malware repository by a user in Bangladesh, the programs are highly customizable. After attackers compromised the targeted banks’ infrastructure and obtained operators credentials, the malware submitted fraudulent messages and rewrote messages sent to printers—or displayed in PDFs—circumventing a primary way that banks audited their money flows.
“The tool was custom-made for this job, and shows a significant level of knowledge of SWIFT Alliance Access software as well as good malware coding skills,” BAE Systems’ researcher Sergei Shevchenko stated in his report on the analysis.
While authorities have not attributed the hack to any particular group, the researchers from BAE Systems have found coding similarities between the malware and the programs used to compromise and disrupt Sony Pictures. The U.S. government has linked that hack to North Korea although some security researchers have expressed doubts about the veracity of their claims.
The theft has led to a great deal of finger-pointing, while at the same time bringing together the Bangladesh central bank, the Federal Reserve Bank of New York and the SWIFT Alliance in a joint investigation.
SWIFT Hack Shows Attackers Penetrating Bank Defenses
Bangladeshi authorities have blamed SWIFT technicians, and officials at the Federal Reserve Bank of New York, for introducing vulnerabilities into their network when they connected the alliance’s real-time gross settlement (RGTS) system, according to a Reuters report. The missteps resulted in the messaging systems at the Bangladesh Bank, the central financial institution for the country, being accessible through a simple password, bank officials told Reuters.
The SWIFT Alliance refuted the arguments, stressing that none of the issues were caused by its systems, but by the insecurity of the Bangladesh Bank’s infrastructure.
“The key defense against such attack scenarios remains for users to implement appropriate security measures in their local environments to safeguard their systems—in particular those used to access SWIFT—against such potential security threats,” the SWIFT Alliance said in an April statement. “Such protections should be implemented by users to prevent the injection of malware into, or any misappropriation of, their interfaces and other core systems.”
The blame game only serves the attackers. The malware, already customized to attack the Bangladesh Bank and another victim institution in Vietnam, will likely be customized to attack more financial houses, according to network-security firm Damballa.
“The success of the attack on Bangladesh Bank’s SWIFT system means that it’s likely we’ll encounter BBSwift [Damballa’s name for the malware] again in a modified form,” the company said in a blog post. “In the next attack, most or all of the local, host-based IOCs (indicators of compromise) are likely to change in an attempt to evade anti-virus and other host-based security controls.”
The SWIFT Alliance could go a long way toward helping its members secure their part of the network by publishing more concrete guidance and requiring that consultants be used to verify a member’s security, Skyport Systems said in an analysis of the recommendations document.
The current security guidelines for SWIFT, for example, do not require that administrators use dedicated desktops to access the service and do not force the most recent and secure version of communications encryption, TLS 1.2, the firm stated. In addition, SWIFT users should be required to segment their network employing a security device to limit access to the administrators’ computers, the company said.
While SWIFT recommends two-factor authentication, they need to also recommend cordoning off administrative credentials from the SWIFT environment, the Skyport analysis stated. Finally, financial institutions should focus on securing their logging services, as attackers will naturally attempt to corrupt or erase the logs to hide their tracks.
Skyport’s Gourlay pointed to two steps that SWIFT could take today that do not require re-engineering banks’ platforms. “Delivering some published guidelines that are actionable, and requiring that auditors come in in the next 60 to 90 days would both go a long way to helping,” he said.
Summing up the issues, SEC Chair Mary Jo White told attendees at the Reuters Financial Regulation Summit in Washington, D.C., that cyber-security has become the biggest problem facing the world’s financial system.
“What we found [in financial institutions], as a general matter so far, is a lot of preparedness, a lot of awareness but also their policies and procedures are not tailored to their particular risks,” she said.