dc.contributor.advisor |
Kabwe, Ruddy |
|
dc.contributor.postgraduate |
Bhengu, Tamkazi Iris |
|
dc.date.accessioned |
2024-02-08T10:26:11Z |
|
dc.date.available |
2024-02-08T10:26:11Z |
|
dc.date.created |
2024-04-25 |
|
dc.date.issued |
2023 |
|
dc.description |
Mini Dissertation (LLM (Tax Law))--University of Pretoria, 2023. |
en_US |
dc.description.abstract |
This research addresses an important, yet largely unexplored area of South African
tax law, being the use of Artificial Intelligence (AI) in the administration of taxes
pertaining to individuals. The digitalisation of the economy through the introduction of
new technologies and business models has placed a huge burden on tax
administrations and challenged their ability and capacity to administer tax laws. Most
tax laws were drafted with human beings and human functions in mind, thus, the
introduction of disruptive technologies that mimic human functions but differ in some
material respects has left existing laws unable to keep up with the digitalisation of the
economy and the impact of disruptive technologies on tax administration. This has
resulted in the need to incorporate new technologies into the tax administration system
in order to keep pace with the rapidly changing tax and economic landscape. These
technologies are Artificial Intelligence (AI) and machine learning (ML).
Artificial intelligence, refers to the use of automation to enhance and develop the
decision-making capabilities of machines, to enable them to replicate human thinking
in order to improve processes. The use of AI has several beneficial applications in tax
administration, including but not limited to (1) enabling the extraction of greater value
from existing data, (2) creating possibilities for the development of novel and
convenient services for taxpayers and, (3) enabling the better management and
address of tax risks such as tax avoidance, evasion and fraud.
Notwithstanding the aforementioned benefits of using AI in tax administration, such
use also carries certain risks, such as (1) the entrenchment of systemic biases which
result in discriminatory practices towards certain taxpayers, (2) threats to privacy
where the rights of taxpayers to have their personal data protected is infringed upon
in the extraction of data for tax administration purposes, and (3) the potential of AI
tools to spread disinformation and lead to a proliferation of inaccuracies which can be
detrimental to both the tax administration and taxpayers.
The aforementioned risks of AI in tax administration require a robust and clear legal
system to manage them. As of the date of submission of this research, the South
African legislature has not yet introduced any laws aimed at addressing the use of AI
in South Africa. It is for this reason that this research analyses the extent to which the
vi
existing legal framework in South Africa is suitable to manage the risks of using AI in
tax administration.
The research is conducted in the form of a literature review. It considers and analyses
the writings of academic authors, legislation, policy documents, case law and various
web sources on the subject of AI in tax administration. A comparison is made between
the Indian legal system and the South African legal system for the determination of
which system addresses the challenges of using AI in tax administration in the best
manner.
The research ultimately concludes that the benefits of using AI in the administration of
taxes pertaining to individuals in South Africa outweigh the risks associated with such
use. Further, the existing South African legal framework sufficiently addresses the
current challenges of using AI in tax administration, while allowing for the extraction of
the benefits associated with the use of AI in tax administration. |
en_US |
dc.description.availability |
Unrestricted |
en_US |
dc.description.degree |
LLM (Tax Law) |
en_US |
dc.description.department |
Mercantile Law |
en_US |
dc.description.faculty |
Faculty of Laws |
en_US |
dc.description.sdg |
None |
en_US |
dc.identifier.citation |
* |
en_US |
dc.identifier.doi |
https://doi.org/10.25403/UPresearchdata.19029833.v2 |
en_US |
dc.identifier.other |
A2024 |
en_US |
dc.identifier.uri |
http://hdl.handle.net/2263/94384 |
|
dc.language.iso |
en |
en_US |
dc.publisher |
University of Pretoria |
|
dc.rights |
© 2023 University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria. |
|
dc.subject |
UCTD |
en_US |
dc.subject |
Algorithm |
en_US |
dc.subject |
Artificial intelligence |
|
dc.subject |
Automated decision-making |
|
dc.subject |
Tax administration |
|
dc.subject |
Taxpayer rights |
|
dc.title |
A critical look at the challenges and benefits of artificial intelligence (AI) in tax administration : A South African perspective |
en_US |
dc.type |
Mini Dissertation |
en_US |