Security automation considered harmful?

  • Authors:
  • W. Keith Edwards;Erika Shehan Poole;Jennifer Stoll

  • Affiliations:
  • Georgia Institute of Technology, Atlanta, GA;Georgia Institute of Technology, Atlanta, GA;Georgia Institute of Technology, Atlanta, GA

  • Venue:
  • NSPW '07 Proceedings of the 2007 Workshop on New Security Paradigms
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

End-users are often perceived as the weakest link in information security. Because of this perception, a growing body of research and commercial activity is focused on automated approaches to security. With these approaches, security decisions are removed from the hands of the users, and are placed instead in systems themselves, or in remote services or organizations that establish policies that are automatically enforced. We contend that although security automation is potentially beneficial in theory, in practice it is not a panacea for end-user information security. A number of technical and social factors mitigate against the acceptance and efficacy of automated end-user security solutions in many cases. In this paper, we present a discussion of the inherent limitations of automating security for end-users. We then discuss a set of design guidelines for choosing whether to automate end-user security systems. We conclude with a set of research directions focused on increasing the acceptance and efficacy of security solutions for end-users.