Security testing for chatbots

Josip Bozic*, Franz Wotawa

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference paperpeer-review


Services like chatbots that provide information to customers in real-time are of increasing importance for the online market. Chatbots offer an intuitive interface to answer user requests in an interactive manner. The inquiries are of wide-range and include information about specific goods and services but also financial issues and personal advices. The notable advantages of these programs are the simplicity of use and speed of the search process. In some cases, chatbots have even surpassed classical web, mobile applications, and social networks. Chatbots might have access to huge amount of data or personal information. Therefore, they might be a valuable target for hackers, and known web application vulnerabilities might be a security issue for chatbots as well. In this paper, we discuss the challenges of security testing for chatbots. We provide an overview about an automated testing approach adapted to chatbots, and first experimental results.

Original languageEnglish
Title of host publicationTesting Software and Systems - 30th IFIP WG 6.1 International Conference, ICTSS 2018, Proceedings
PublisherSpringer Verlag Heidelberg
Number of pages6
ISBN (Print)9783319999265
Publication statusPublished - 1 Jan 2018
Event30th IFIP International Conference on Testing Software and Systems, ICTSS 2018 - Cadiz, Spain
Duration: 1 Oct 20183 Oct 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11146 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference30th IFIP International Conference on Testing Software and Systems, ICTSS 2018


  • Adaptive systems
  • chatbots
  • security testing

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)


Dive into the research topics of 'Security testing for chatbots'. Together they form a unique fingerprint.

Cite this