A Logic for the Compliance Budget

Gabrielle Anderson, Guy McCusker, David Pym

Research output: Chapter or section in a book/report/conference proceedingChapter or section

3 Citations (SciVal)
198 Downloads (Pure)


Security breaches often arise as a result of users? failure to comply with security policies. Such failures to comply may simply be innocent mistakes. However, there is evidence that, in some circumstances, users choose not to comply because they perceive that the security benefit of compliance is outweighed by the cost that is the impact of compliance on their abilities to complete their operational tasks. That is, they perceive security compliance as hindering their work. The ?compliance budget? is a concept in information security that describes how the users of an organization?s systems determine the extent to which they comply with the specified security policy. The purpose of this paper is to initiate a qualitative logical analysis of, and so provide reasoning tools for, this important concept in security economics for which quantitative analysis is difficult to establish. We set up a simple temporal logic of preferences, with a semantics given in terms of histories and sets of preferences, and explain how to use it to model and reason about the compliance budget. The key ingredients are preference update, to account for behavioural change in response to policy change, and an ability to handle uncertainty, to account for the lack of quantitative measures.
Original languageEnglish
Title of host publicationProceedings, GameSec 2016- Decision and Game Theory for Security
Subtitle of host publication7th International Conference, GameSec 2016, New York, NY, USA, November 2-4, 2016
EditorsQuanyan Zhu, Tansu Alpcan, Emmanouil Panaousis, Milind Tambe, William Casey
PublisherSpringer Verlag
ISBN (Print)9783319474120
Publication statusPublished - 2016

Publication series

NameLecture Notes in Computer Science


Dive into the research topics of 'A Logic for the Compliance Budget'. Together they form a unique fingerprint.

Cite this