Algorithmic automation in criminal proceedings: A general discussion and Czech perspective
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11220%2F25%3A10502456" target="_blank" >RIV/00216208:11220/25:10502456 - isvavai.cz</a>
Result on the web
<a href="https://doi.org/10.4324/9781003531500-17" target="_blank" >https://doi.org/10.4324/9781003531500-17</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.4324/9781003531500-17" target="_blank" >10.4324/9781003531500-17</a>
Alternative languages
Result language
angličtina
Original language name
Algorithmic automation in criminal proceedings: A general discussion and Czech perspective
Original language description
This paper addresses the utilisation of algorithms in the automation of criminal proceedings, with a particular focus on the decision-making process. The author presents a critical perspective on the integration of computer-run algorithms into the decision-making process, proposing that the human factor, based on the recognition of personal experience, rather than a purely statistical approach, is a fundamental pillar of the criminal justice system. The author identifies the primary challenges associated with the utilisation of contemporary artificial intelligence (AI) tools in criminal proceedings across the globe. He illustrates this with a case study of the "COMPAS" risk assessment tool, which has been adopted by multiple jurisdictions in the USA for the estimation of the risk of recidivism for a given defendant. The author posits that these tools face a significant challenge because of the opacity of the algorithmic process. This is due to the fact that the software is typically subject to the trade secrets of private technology companies, while the details of the data processing procedure are not publicly available. Furthermore, this paper posits that even judges and law enforcement agencies lack comprehensive knowledge of the precise operational background of these systems. Nevertheless, they continue to utilise the outputs of the algorithms as a foundation for their decisions. Furthermore, the author addresses the issue of potential bias amplification by software tools. This can occur when statistically relevant discriminatory elements from previous decisions are treated non-critically by the algorithm, as if they were a regular part of the given dataset. The author addresses the potential utilisation of algorithmic decision-making in the context of Czech criminal procedural law, given that it has not yet been implemented in any form within Czech criminal proceedings. Additionally, the author addresses the resolution concerning the deployment of artificial intelligence in criminal proceedings within the European context.
Czech name
—
Czech description
—
Classification
Type
C - Chapter in a specialist book
CEP classification
—
OECD FORD branch
50501 - Law
Result continuities
Project
—
Continuities
S - Specificky vyzkum na vysokych skolach
Others
Publication year
2025
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Book/collection name
The use of algorithms in criminal law: A Central European Perspective
ISBN
978-1-03-287183-7
Number of pages of the result
14
Pages from-to
151-164
Number of pages of the book
240
Publisher name
Routledge
Place of publication
New York
UT code for WoS chapter
—