Recently, a group called Shadow Brokers released hundreds of megabytes of tools claimed to be stemming from the NSA and other intelligence organizations. Ars has written extensively on the subject: https://arstechnica.com/security/2017/04/nsa-leaking-shadow-brokers-just-dumped-its-most-damaging-release-yet/. The leaked code is available on github.com/misterch0c/shadowbroker. The exploits target several Microsoft products still in service (and commonly used), as well as the SWIFT banking network. Adding speculation to the case is the fact that Microsoft silently released patches to vulnerabilities claimed to be zerodays in the leaked code prior to the actual leak. But what does all of this mean for “the rest of us”?
There are two key questions we need to ask and try to answer:
- Should threat models include domestic nation state actors, including illegal use of intelligence capabilities against domestic targets?
- Does the availability of the leaked tools increase the threat from secondary actors, e.g. organized crime groups?
Taking on the first issue first: should we include domestic intelligence in threat models for “normal” businesses? Let us examine the C-I-A security triangle from this perspective.
- Confidentiality: are domestic intelligence organizations interested in stealing intellectual property or obtaining intelligence on critical personnel within the firm? This tends to be either supply chain driven if you are not yourself the direct target, or data collection may occur due to innocent links to other organization’s that are being targeted by the intelligence unit.
- Integrity (data manipulation): if your supply chain is involved in activities drawing sufficient attention to require offensive operations, including non-cyber operations, integrity breaches are possible. Activities involving terrorism funding or illegal arms trade would increase the likelihood of such interest from authorities.
- Availability: nation state actors are not the typical adversary that will use DoS-type attacks, unless it is to mask other intelligence activities by drawing response capabilities to the wrong frontier.
The probability of APT activities from domestic intelligence is for most firms still low. The primary sectors where this could be a concern are critical infrastructure and financial institutions. Also firms involved in the value chains of illegal arms trade, funding of terrorism or human trafficking are potential targets but these firms are often not aware of their role in the illegal business streams of their suppliers and customers.
The second question was if the leak poses an increased threat from other adversary types, such as organized crime groups. Organized crime groups run structured operations across multiple sectors, both legal and illegal. They tend to be opportunistic and any new tools being made available that can support their primary cybercrime activities will most likely be made use of quickly. The typical high-risk activities include credit card and payment fraud, document fraud and identity theft, illicit online trade including stolen intellectual property, and extortion schemes by direct blackmail or use of malware. The leaked tools can support several of these activities, including extortion schemes and information theft. This indicates that the risk level does in fact increase with the leaks of additional exploit packages.
How should we now apply this knowledge in our security governance?
- The tools use exploits in older versions of operating systems. Keeping systems up-to-date remains crucial. New versions of Windows tend to come with improved security. Migration prior to end-of-life of previous version should be considered.
- In risk assessments, domestic intelligence should be considered together with foreign intelligence and proxy actors. Stakeholder and value chain links remain key drivers for this type of threat.
- Organized crime: targeted threats are value chain driven. Most likely increased exposure due to new cyberweapons available to the organized crime groups for firms with exposure and old infrastructure.