When will my PLC support Mirai? The security economics of large-scale attacks against internet-connected ICS devices

Dodson, Michael and Beresford, Alastair R. and Thomas, Daniel R.; (2021) When will my PLC support Mirai? The security economics of large-scale attacks against internet-connected ICS devices. In: Proceedings of the 2020 APWG Symposium on Electronic Crime Research, eCrime 2020. eCrime Researchers Summit, eCrime . IEEE, Piscataway, N.J.. ISBN 9781665425391 (https://doi.org/10.1109/eCrime51433.2020.9493257)

[thumbnail of Dodson-etal-APWG-eCrime-2020-When-will-my-PLC-support-Mirai-the-security-economics-of-large-scale-attacks]
Preview
Text. Filename: Dodson_etal_APWG_eCrime_2020_When_will_my_PLC_support_Mirai_the_security_economics_of_large_scale_attacks.pdf
Accepted Author Manuscript

Download (176kB)| Preview

Abstract

For nearly a decade, security researchers have highlighted the grave risk presented by Internet-connected Industrial Control Systems (ICS). Predictions of targeted and indiscriminate attacks have yet to materialise despite continued growth of a vulnerable population of devices. We investigate the missing attacks against ICS, focusing on large-scale attacks enabled by Internet-connected populations. We fingerprint and track more than 10,000 devices over four years to confirm that the population is growing, continuously-connected, and unpatched. We also track 150,000 botnet hosts, monitor 120 global ICS honeypots, and sift 70 million underground forum posts to show that the cybercrime community has little competence or interest in the ICS domain. Attackers may be dissuaded by the high cost of entry, the fragmented ICS population, and limited onboard resources; however, this justification is incomplete. We use a series of case studies to develop a security economics model for large-scale attacks against Internet-connected populations in general, and use it to explain both the current lack of interest in ICS and the features of Industry 4.0 that will make the domain more accessible and attractive to attackers.