Criminal liability of autonomous agents: from the unthinkable to the plausible

Pedro Miguel Freitas*, Francisco Andrade, Paulo Novais

*Corresponding author for this work

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

6 Citations (Scopus)


The evolution of information technologies have brought us to a point where we are confronted with the existence of agents - computational entities-which are able to act autonomously with little or no human intervention. And their behavior can damage individual or collective interests that are protected by criminal law. Based on the analysis of different models of criminal responsibility of legal persons - which constituted an interesting advance in the criminal law in relation to what was hitherto traditionally accepted -, we will appraise whether the necessary legal elements to have direct criminal liability of artificial entities are present.
Original languageEnglish
Title of host publicationAI approaches to the complexity of legal systems
EditorsPompeu Casanovas, Ugo Pagallo, Monica Palmirani, Giovanni Sartor
Place of PublicationHeidelberg
PublisherSpringer-Verlag Berlin Heidelberg
Number of pages12
ISBN (Electronic)9783662459607
ISBN (Print)9783662459591
Publication statusPublished - 2014
Externally publishedYes
EventAICOL 2013 International Workshops: AICOL-IV@IVR - Belo Horizonte, Brazil
Duration: 21 Jul 201327 Jul 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
ISSN (Print)0302-9743


ConferenceAICOL 2013 International Workshops
CityBelo Horizonte


  • Autonomous agents
  • Criminal liability
  • Legal persons
  • Objects
  • Software agents


Dive into the research topics of 'Criminal liability of autonomous agents: from the unthinkable to the plausible'. Together they form a unique fingerprint.

Cite this