A threat actor calling itself “Bozkurt Hackers” posted links to data on Twitter allegedly obtained from a number of banks based in the United Arab Emirates, Bangladesh and Nepal. Posts made by Twitter accounts affiliated with Bozkurt Hackers in relation to the claimed data leaks referenced a number of journalists, media outlets and security researchers. The accounts shared similar, related imagery, names and the subject matter relevance of the posts – specifically in relation to data allegedly pertaining to the banks in question.

Bozkurt Hackers first emerged when in April when it claimed responsibility for the Qatar National Bank (QNB) compromise; the group posted QNB data on the Cryptome website on 25 April 2016. Twitter accounts affiliated with Bozkurt Hackers subsequently claimed that the group would release additional unspecified data at various points on, and after, 27 April 2016.

On 06 May 2016, a Twitter affiliated with the group (@bozkurt_1923_) provided a download link to approximately 10GB data that pertained to the Sharjah-based Invest Bank. Invest Bank publicly stated that this data was from a previous confirmed data breach in 2015. This was consistent with the dates included in some of the files in the Invest Bank data released by Bozkurt Hackers. A lesser amount of the same data was originally leaked by an extortion-motivated actor called “Hacker Buba” in November 2015. Additionally, we detected reference to Hacker Buba on a clear web carding forum by a user named “Bozkurt” on 28 March 2016 that provided a link to Invest Bank data relating to the “cashing out” of Arab bank. This post was added to other criminal forums and a Twitter account affiliated with Bozkurt Hackers (@ulkuocaklar1923) referenced these posts.

Bozkurt carding

Figure 1 – Reference to Hacker Buba by forum user “Bozkurt”

More recently, on 5 May 2016, the Twitter account @bozkurt_turk_ posted a screenshot of a number of folders named after countries and continents including; Syria, Malaysia, Iran, India, Greece, Australia, Africa, Bahrain, Bangladesh & Nepal, Europe, The Philippines, Qatar, Russia, United Arab Emirates and the United States of America.

Bozkurt Hackers Folders

Figure 2 – Screenshot of folders provided by Bozkurt Hackers

On 9 May, the Twitter account @turkey_bozkurt posted a number of download links that allegedly pertained to data obtained from three banks based in Bangladesh and two banks based in Nepal. While we observed that the size of the archives available for download from the alleged Nepal and Bangladesh banks were limited (approximately 500MB and below), one .CSV file included 326,747 rows of data.

The exact motivation of Bozkurt Hackers is uncertain and it was not known if the actor leaking the data was responsible for the initial intrusions. The links between the user on the carding forum demonstrated in Figure 1 show that it is possible the threat actor motivations include financial gain. Although their method of releasing data freely and publicly may appear contrary to these aims, older data that has already been used may be of limited value for fraud; therefore it is possible that this has been used to gain notoriety. Given that the threat actor has demonstrably contacted media outlets and journalists, it is likely that a significant facet of their motivation is also to gain publicity for their activity. The imagery used by the threat actor on their Twitter accounts shows some affiliation with the ideology of the Turkish Nationalist group called Ülkü Ocakları (Grey Wolves). Ülkü Ocakları seek to promote Turkish nationalism over Arab nationalism, and although the attribution of the “Bozkurt Hackers” to the compromise of the Qatar National Bank is highly questionable, this ideology does provide some possible explanation for their claims regarding that data leakage.

Based on this continued trend of data leakage by affiliated actors, coupled with the fact the group followed up on threats to release additional data in the past, it is a realistic possibility that the group will leak further data. Bozkurt Hackers have demonstrated a willingness to engage with journalists and media outlets. There is, therefore, a realistic possibility that further insight will emerge into the group’s activity through these sources.