Use of artificial intelligence by tax administrations: An analysis regarding taxpayers’ rights in Latin American countries

doi:10.1016/j.clsr.2020.105441

Computer Law & Security Review

Volume 38, September 2020, 105441

https://doi.org/10.1016/j.clsr.2020.105441 Get rights and content

Abstract

In this paper, we analyze taxpayers’ rights to have access to artificial intelligence algorithms and formulas that have been used by tax administrations in Latin America. We consider two applications of artificial intelligence: in the characterization of taxpayers’ risk and the robotization of tax audit actions. Very little has been described in the literature on how these technologies coexist with taxpayers’ rights, especially in the exercise of their right to defense in administrative and contentious proceedings. The evidence reflects that, although in the countries under study the access to these techniques is not clearly regulated, general principles derived from the fundamental rights declared by each country make it possible to safeguard taxpayers’ right to access this information.

Section snippets

Inroduction

Tax administrations have the main function of managing tax compliance to detect and prevent criminal behavior and provide service and education to help taxpayers meet their tax obligations with the least complexity and burden of compliance.¹

General aspects of technological systems used by tax administrations

New technologies have led to critical changes in international politics, reducing information collection costs, decreasing market friction and significantly driving the process of world market expansion.⁷ In

Comparison of taxpayers’ rights in Latin America countries

Taxpayers’ rights have been defined as the obligations or duties that the State must fulfill, through its tax administration office, in terms of tax collection, control and education functions.³⁶ When these rights are

Techniques for characterization of taxpayers’ risk

Techniques that identify patterns of taxpayers' behavior have been key to detecting fraud or tax evasion, also allowing measurements of the risk of compliance with tax obligations. For this purpose, tax administrations have focused on the use of business intelligence, defined as a set of information systems that support decision-making, based on other storage, analysis and data extraction technologies.⁴²

Robotization techniques in tax audit actions

Watson, an AI computer system developed in 2011 by International Business Machines Corp. (IBM), revolutionized the professional practice of management and advice of public and private companies. Watson was the most important technology to arrive in the field of law and allowed professionals to think innovatively.⁷⁴

Conclusions

Having shown that the AI mechanisms used by some tax administrations have normative implications in social life, it is essential to specify the legal limits of their application in safeguarding taxpayers’ rights, regardless of the way in which these are enshrined in the legal system, and the regulatory level at which their guarantee is found.

Considering this, the characterization techniques used by Latin American tax administrations, such as Chile, Peru, Brazil, Colombia, Ecuador, Argentina and

References (0)

Cited by (27)

Digital technologies, artificial intelligence, and bureaucratic transformation
2022, Futures
Citation Excerpt :
In areas like tax collection, criminal justice, and public health, sophisticated computerized data processing systems are becoming essential elements in the implementation of public policy and the delivery of public services. For example, advanced computerized data processing has been used in Brazil to control tax evasion (Faúndez-Ugalde et al., 2020: 3), in the United States to provide guidance on whether to hold or release a defendant before a criminal trial is held (Rizer & Watney, 2018), and in Singapore to aid in contact tracing as part of the COVID-19 pandemic response (Goggin, 2020). These automated systems save time, reduce human capital costs, free up personnel to work on other tasks, curb biases and discrimination, and enable the consideration of problems too complex for human analysts.
Bureaucracies are often criticized for their inflexibility, budget-maximizing wastefulness, and excessive rules and procedures. Rapid advances in technology, including the expansion of digital government, the use of artificial intelligence, and the ability to collect and analyze big data, promise to make public sector organizations leaner, more efficient, and more responsive to citizens' needs. While these technological changes have prompted some observers to forecast the end of bureaucracy, data from many countries show that bureaucratic public organizations are not disappearing. In this article, we argue that this paradox can be explained by revisiting some of the foundational work of sociologist Max Weber, who envisioned public administration itself as a bureaucratic machine. Advanced computing technologies, like artificial intelligence, are reinforcing bureaucratic tendencies in the public sector, not eliminating them. While advances in technology may transform the way public sector organizations operate, they can also serve to strengthen bureaucracy's core purpose.
Trends, impacts, and prospects for implementing artificial intelligence technologies in the energy industry: The implication of open innovation
2021, Journal of Open Innovation: Technology, Market, and Complexity
Citation Excerpt :
AI creates surplus value for businesses because it helps to use the full potential of available data, makes reliable forecasts, and automates complex tasks. It also enhances productivity by automating business processes and targets that humans previously implemented [18,36,37,38]. Many Russian energy companies are now improving growth rates and productivity with AI solutions [8,39,40].
This research aims to substantiate the impact of using open innovation (OI) in the energy sector in readiness to implement artificial intelligence (AI) technologies and their effectiveness. The empirical method was proposed to determine the readiness level of OI for the implementation of AI technologies by comparing Russian and French energy companies. Readiness level indicators of companies for AI implementation using the Fibonacci sequence, Student’s t-test, and the method of fuzzy sets were empirically determined. The integrated readiness indicator for AI implementation by companies was calculated using the method of fuzzy sets and expressed through variance, allowing for these significant factors. Russian companies are at a low level of developmental readiness to implement AI, which is in contrast to companies operating in a developed market where the determining factor is the AI technology cost. The example of the innovative business model “Energy-as-a-Service” shows the synergistic effects of OI use and AI technology introduction. This paper is novel because it seeks to contribute to the current debate in the literature, justifying the position that energy companies that have in the past actively applied the concept of open innovation in business, are the most competitive and most efficient in implementing AI technologies.
Cryptocurrencies, Tax Ignorance and Tax Noncompliance in Direct Taxation: Spanish Empirical Evidence
2024, Economies
Artificial intelligence research in Nigeria: Topic modelling and scientometric analysis
2024, IAES International Journal of Artificial Intelligence
Leveraging the potential of artificial intelligence (AI) in exploring the interplay among tax revenue, institutional quality, and economic growth in the G-7 countries
2024, AI and Society
Digital taxation, artificial intelligence and Tax Administration 3.0: improving tax compliance behavior – a systematic literature review using textometry (2016–2023)
2024, Accounting Research Journal

View all citing articles on Scopus

: Article developed within the internal research project of the Vice-Rector of Research and Advanced Studies of the Pontificia Universidad Católica de Valparaíso, COD. PROJECT: 039.406/2019.

View full text