In the 2017 Annual Threat Report, the eSentire Threat Intelligence team documented a series of scenarios that could potentially occur though 2018. The likelihood of each forecast scenario increased or decreased according specific indicator events occurring. The 2017 report also contained a trend analysis section with five trends from 2017 that accurately carried through 2018. With 2018 now complete, this blog post reviews the list of scenarios and indicators to assess the 2018 forecast accuracy.
2018 Scenarios List
- Cyber criminals adopt swarm methods for command and control execution to harden blocking effort
- Cyber criminals adopt machine learning to increase the scale of successful exploitations
- Cyber criminals give preference to crypto mining malware resulting in pandemic threat increase
- Cyberwarfare methods are adopted and used by radical groups to cause damage to private organizations
- Cyber-attacks are used to cause physical damage infrastructure objects and/or loss of lives
- Increased information stealing attacks targeting crypto wallets
- Cybercriminals focus attention on attacking/exploiting mobile devices, IoT and smartphones over personal computers
For a full explanation of each scenario and the correlating indicators, please see the 2017 Annual Threat Report [1].
Verified Forecasts
From this list, there were three scenarios that were positively confirmed through 2018.
- Cyber criminals give preference to crypto mining malware resulting
- Increased information stealing attacks targeting crypto wallets
- Cybercriminals focus attention on attacking/exploiting mobile devices, IoT and smartphones
2018 saw major increases in threat actor activity targeting cryptocurrencies; both through illicit mining [2] and targeted attacks on cryptocurrency wallets [3]. In late January 2018, eSentire detected a supply chain attack exploiting Kasaya’s Virtual System Administrator (VSA) to deliver cryptocurrency miners; this major event set a trend for 2018 [4]. One of the interesting aspects surrounding this rise is that the targeting of cryptocurrencies did not decrease as cryptocurrency values declined. This is likely due to a variety of contributing reasons, including the comparable simplicity of monetizing illicitly gained cryptocurrencies and the perceived anonymity of cryptocurrencies.
Throughout 2018, eSentire detected a high amount of IoT exploitation attempts, in line with an increase of cyber criminal attention on IoT devices. The largest spikes specifically targeted cameras, door controllers, surveillance equipment and media devices. Events that indicated attackers increased focus on IoT devices include the continuing low awareness of individuals relating to IoT devices, the ever increasing number of deployed devices and wormable IoT malware [5]. IoT device compromise is generally designed for indirect financial gain; compromising devices that can then be used in later attacks to generate illicit revenue.
Out of the seven scenarios from the 2017 annual report, Cyber criminals adopt swarm methods for command and control execution to harden blocking effort, is the only scenario to fall in between the generic did or did not categories. Peer to peer botnets have been used by threat actors for some time now; these botnets are structured in a decentralized way and do not require a standard command and control infrastructure. This botnet structure helps criminals to avoid a full takedown of services by law enforcement. Through 2018 we have not seen the advanced aspect of a swarm botnet, where each bot acts as an individual intelligent piece of a larger cluster, capable of discovering vulnerable systems and targets without specific instruction being passed down. Botnets continue to evolve and pose a threat, but true swarm technology botnets have not yet been identified in the wild.
Forecasts that did not Occur
The following three scenarios were not confirmed through 2018:
- Cyber criminals adopt machine learning to increase the scale of successful exploitations Cyberwarfare methods are adopted and used by radical groups to cause damage to private organizations
- Cyber-attacks are used to cause physical damage infrastructure objects and/or loss of lives
The indicators tracked for the above three scenarios were rarely seen or were negative events. Machine learning continues to be improved and applied in a real world context, but attacks in the wild employing machine learning have yet to be seen. Cyberwarfare is a vague term and open to some interpretation, but the known cyber operations used by radical groups has remained limited. Attacks by radical groups in 2018 have been primarily defacement and recruitment. Lastly, the potential that cyber-attacks cause physical infrastructure damage or loss of life remains real; in early 2018 the Triton ICS malware was publicly identified in a real world attack and had the potential to cause major damage and loss of life but this outcome was luckily avoided [6]. Although each scenario remains plausible, the likelihood of occurrence in the near future is limited.
A Final Word
By their nature, security predictions are unstable; any number of events can dramatically change the threat landscape and threat actor tactics. By identifying and tracking indicators that raise or lower the likelihood of a potential scenario it becomes possible to better know what to expect. For a complete security focused over-view of 2018, including industry trends and threat data, see the eSentire 2018 Annual Threat Report.
