Intelligent Tax Systems: Automating Tax Audits and Improving Revenue Efficiency

The introduction of Intelligent Tax Systems (ITSs), driven by Artificial Intelligence (AI) and Machine Learning (ML), is transforming tax administration by automating routine tasks, enhancing audit processes, and improving overall revenue collection efficiency. Traditional tax audits, reliant on manual data checks and human intervention, are slow and prone to errors. ITS can analyze vast amounts of tax data in real time, detect discrepancies, predict potential fraud, and automate compliance processes. These advancements significantly improve the speed, accuracy, and transparency of tax audits. Additionally, ITS can optimize tax revenue collection by identifying high-risk cases and allocating resources more efficiently. Despite its many benefits, the implementation of ITS comes with challenges, such as data privacy concerns, system integration issues, and regulatory compliance. This paper explores the potential of Intelligent Tax Systems in automating tax audits and improving revenue efficiency, examines the current state of their application, and discusses the challenges faced during implementation. The paper highlights the future prospects of ITS and its role in modernizing tax administration.

• Intelligent Tax Systems
• Tax Automation
• Revenue Efficiency
• Fraud Detection
• Machine Learning
• Tax Audits
• AI
• Data Analytics
• Tax Compliance
• Automation

1. Introduction

Tax systems are the cornerstone of a country’s financial infrastructure, ensuring that governments have the necessary resources to fund public services, infrastructure, and development projects ( Cahyadi, 2024 ). The increasing complexity of global financial activities, however, has made traditional tax administration processes less effective ( Nurbekova, Juchnevicius, & Tussupova, 2024 ). Historically, tax audits have been time-consuming and heavily dependent on human expertise, relying on manual checks of financial records, tax returns, and discrepancies. With the rapid advancement of technology, particularly Artificial Intelligence (AI) and Machine Learning (ML), tax authorities now have the opportunity to revolutionize the way audits are conducted and improve revenue efficiency ( Rahman et al., 2024 ).

Intelligent Tax Systems (ITSs) refer to AI-driven platforms that automate tax audits, improve fraud detection, and optimize revenue collection by utilizing advanced algorithms and data analytics ( Ariyibi et al., 2024 ). These systems can analyze massive amounts of taxpayer data, detect patterns, and identify anomalies that may indicate tax evasion or fraud. Unlike traditional manual audits, which are often inefficient and prone to errors, ITS provides a faster, more accurate, and automated approach to tax administration ( Ezeife et al., 2021 ). For example, AI can examine financial transactions in real-time, identify irregularities, and flag them for further investigation. This proactive approach reduces the time required to perform audits and minimizes human error, significantly increasing the efficiency of tax administration.

In addition to automating audits, ITS also plays a crucial role in enhancing revenue collection efficiency ( Kirer, 2024 ). Traditional tax systems often rely on manual or random selection for audits, which can result in wasted resources and missed opportunities for tax revenue generation. By leveraging machine learning, ITS can prioritize audits based on data-driven assessments of taxpayer risk. For instance, AI systems can assess the likelihood of fraud based on historical data, behavior patterns, and inconsistencies in reported financial data, allowing tax authorities to focus their resources on the highest-risk cases. This targeted approach ensures that revenue is collected more efficiently and accurately, improving the overall fiscal health of the government.

Another advantage of ITS is its potential to improve taxpayer compliance. As tax regulations become increasingly complex, many taxpayers struggle to understand their obligations and navigate the system. Intelligent tax systems can provide real-time support by guiding taxpayers through the filing process, sending reminders, and offering explanations for tax requirements ( Çetin Gerger, 2019 ). This reduces the likelihood of errors, misunderstandings, and unintentional non-compliance, ultimately promoting greater adherence to tax laws.

Despite the clear advantages, the implementation of ITS is not without its challenges. One of the primary concerns is data privacy. Tax data is highly sensitive, and the use of AI to process this information raises questions about the security and confidentiality of taxpayers’ personal and financial data. Tax authorities must ensure that AI systems comply with privacy regulations, such as the General Data Protection Regulation (GDPR) in Europe and the California Consumer Privacy Act (CCPA), to safeguard taxpayer information. Additionally, system integration remains a challenge. Many tax agencies still rely on outdated infrastructure, and integrating modern AI systems with existing platforms can be complex and costly ( Ezeife et al., 2025 ). Lastly, there are concerns about bias in AI algorithms. If AI models are trained on incomplete or biased data, they can perpetuate existing disparities, leading to unfair treatment of certain taxpayers. Ensuring that ITS is transparent, fair, and free from bias is essential for maintaining trust in the system.

This paper examines the role of Intelligent Tax Systems in modernizing tax administration by automating tax audits and improving revenue efficiency. It explores the capabilities of AI-driven systems, particularly in fraud detection and tax compliance, and assesses the challenges involved in implementing these systems. Additionally, the paper discusses the future potential of ITS in reshaping the tax landscape, highlighting the need for technological advancements and regulatory considerations to ensure the successful adoption of AI-driven tax systems.

2. Methodology

This paper employs a mixed-method approach combining a systematic literature review with empirical data analysis from multiple sources. The quantitative data presented in Figures 1-3 , Table 1 and Table 2 were compiled from three primary sources: 1) IMF Digital Tax Implementation reports covering 189 countries from 2020-2024, 2) longitudinal case studies from the Georgian tax reform (2004-2011) and Ethiopian Electronic Sales Reporting Mechanism (ESRM) implementation (2020-2023), and 3) performance metrics from Thomson Reuters AI solutions deployed across 15 tax jurisdictions.

Data collection involved systematic extraction of implementation success rates, performance metrics, and comparative analyses from official government reports, peer-reviewed publications, and verified industry case studies. The Georgian case study data spans seven years of tax-to-GDP ratio improvements, while the Ethiopian ESRM data covers a three-year implementation period with sample sizes of 12,000+ registered businesses. Success rates were calculated using weighted averages across jurisdictions, with confidence intervals of 95% for reported percentages. Limitations include potential reporting bias from self-selected case studies and varying implementation contexts across countries.

3. The Need for Intelligent Tax Systems

Traditional tax systems have long been criticized for their inefficiency and lack of accuracy. These systems often rely heavily on manual audits, random selection, and error-prone processes to detect fraud and ensure compliance, which limits their effectiveness and creates significant delays in tax enforcement ( Bezditnyi, 2024 ). With the rapid growth of digital financial transactions and the increasing complexity of global financial networks, these traditional systems are ill-equipped to keep pace with modern economic activity. Furthermore, financial data is growing exponentially, with companies generating large amounts of complex data from numerous sources ( Tomar & Periyasamy, 2023 ). Handling such vast amounts of data with outdated methods is no longer feasible.

Figure 1 demonstrates the superior performance of intelligent tax systems across six key operational dimensions, based on empirical studies such as Fedyk et al. (2022) and multi-country implementation data compiled by the IMF.

Figure 1

Figure 1. Traditional vs intelligent tax system audit process comparison.

ITS aims to address these challenges by leveraging AI and ML to automate and enhance various aspects of tax administration ( Scientific, 2024 ). AI allows these systems to analyze vast datasets quickly and accurately, identifying patterns and detecting discrepancies that would otherwise go unnoticed. For example, by processing tax returns, financial records, and publicly available data, ITS can flag discrepancies, detect anomalies in reported income, and identify suspicious financial activities in real time. This automated analysis significantly reduces the time needed for audits, as tax authorities can now focus on high-risk cases rather than conducting routine checks.

Furthermore, ITS can improve fraud detection by using advanced algorithms that predict patterns of evasion and flag potential fraud before it happens. Unlike traditional audit systems that operate reactively, AI-powered systems are proactive, enabling tax authorities to detect and stop fraudulent activities early in the process. By analyzing historical tax data and transaction trends, ITS can predict suspicious behavior or patterns that deviate from typical financial activities, such as unreported income, false deductions, or tax avoidance schemes. This predictive capability allows tax authorities to allocate resources efficiently, minimizing the chances of missing fraudulent activities and maximizing the recovery of evaded taxes ( Olabanji et al., 2024 ).

Table 1 presents documented implementation success rates from real-world government deployments, with data collection modules achieving 92% success rates across 189 countries studied by the IMF, while pattern recognition engines show 85% success based on the Armenian gradient boosting implementation.

Table 1

Table 1. Key components of intelligent tax systems.

Component	Primary Function	Technology Used	Implementation Difficulty	Success Rate (%)	Source Study
Data Collection Module	Gather Taxpayer Information	APIs, Web Scraping	Low	92	IMF Digital Tax
Pattern Recognition Engine	Detect Anomalies and Fraud	Machine Learning	Medium	85	Armenian Tax Study
Risk Assessment Tool	Prioritize Audit Cases	Predictive Analytics	Medium	78	US IRS Program
Real-Time Monitoring	Continuous Surveillance	Stream Processing	High	71	Ethiopia ESRM Study
Compliance Assistant	Guide Taxpayer Filing	NLP, Chatbots	Low	89	Digital Filing Systems
Reporting Dashboard	Visualize Insights	Business Intelligence	Low	94	Multiple Implementations

The challenge of detecting fraud is compounded by the increasing complexity of financial transactions. Businesses engage in a wide range of transactions across global markets, involving numerous financial instruments, subsidiaries, and cross-border dealings. Traditional manual auditing processes struggle to keep up with the intricacies of modern financial transactions ( Đukić, Pavlović, & Grdinić, 2023 ). ITS can overcome this limitation by processing and analyzing large datasets at speed, identifying risks, and making real-time decisions to detect inconsistencies or illegal activities.

In addition to fraud detection, ITS significantly improves revenue efficiency by automating the collection of tax data, improving compliance, and streamlining resource allocation. Traditional tax systems often suffer from inefficiencies in identifying tax revenue shortfalls or delays in collections ( Okunogbe & Santoro, 2023 ). With intelligent systems, tax authorities can optimize tax collection by focusing on high-risk cases identified by AI models, ensuring that resources are allocated where they are most needed. Predictive analytics allow for better forecasting of future tax revenues, providing governments with more accurate and timely insights into their fiscal health. This allows for improved planning, budgeting, and decision-making at the governmental level, ensuring that tax policies are based on the most current and accurate financial data.

4. How Intelligent Tax Systems Work?

At the heart of ITS is the use of advanced machine learning algorithms and data analytics to automate and optimize tax-related processes ( Belahouaoui & Attak, 2024 ). ITS operate by gathering vast amounts of financial data from various sources, including tax filings, business transactions, financial statements, and publicly available data. The system uses machine learning models to analyze these data streams in real-time, detecting anomalies and inconsistencies that could suggest fraud or tax evasion ( Hilal, Gadsden, & Yawney, 2022 ).

The operational workflow of Intelligent Tax Systems follows a systematic four-stage process, as depicted in Figure 2 . This streamlined approach enables continuous monitoring and rapid response to potential tax compliance issues, transforming reactive audit practices into proactive revenue protection strategies.

Figure 2

Figure 2. Intelligent tax system workflow process.

The process begins by collecting taxpayer data from multiple sources. For instance, a tax authority may gather tax returns, accounting records, bank transactions, and sales data. These data are then fed into machine learning models designed to analyze the relationships between different data points. For example, an AI system may examine an individual’s or business’s reported income and compare it to their spending patterns, assets, or industry averages. The system will flag any discrepancies for further investigation.

One of the key features of ITS is real-time monitoring. Unlike traditional tax audits, which can take months or even years to complete, ITS can continuously monitor taxpayer data and immediately identify suspicious activity ( Ruan et al., 2019 ). For example, if a business reports a large decrease in revenue compared to the previous year, the system may flag the discrepancy and trigger an automatic audit. Additionally, predictive analytics is an essential aspect of ITS. By using historical data, the system can predict potential risks, such as unreported income, tax avoidance, or inconsistencies in tax filings, and flag them for review ( Zheng et al., 2024 ). This predictive capacity allows tax authorities to proactively manage fraud and evasion.

AI-driven fraud detection is one of the most critical applications of ITS ( Bello et al., 2023a ). Machine learning models are trained to detect fraudulent patterns by analyzing past instances of fraud, as well as patterns of tax evasion that may emerge in the future. These systems are continuously updated to learn from new data and improve their fraud detection capabilities ( Bello et al., 2023b ). For example, if a taxpayer consistently reports expenses that are disproportionately high compared to their income or industry averages, the system will identify this as a potential red flag and request a deeper investigation.

Additionally, ITS can be used to optimize tax revenue collection. Traditional tax collection systems often suffer from inefficiencies due to outdated methods and incomplete data. With AI-powered systems, tax authorities can predict revenue shortfalls and allocate resources to areas where revenue generation is lagging ( Folorunso & Nwankwo, 2024 ). By leveraging data and predictive models, ITS can help governments collect taxes more efficiently, reducing the time it takes to process payments and increasing overall compliance.

Figure 3 illustrates the substantial increase in government tax revenues following Georgia’s 2004 tax reforms. It shows consistent growth across all major tax categories—particularly in Value Added Tax (VAT), which rose sharply from around 600 million to over 3500 million Georgian Lari. Income and profit taxes also experienced significant increases, indicating enhanced compliance and economic expansion. The chart reflects the positive fiscal impact of simplification, rate reductions, and improved enforcement introduced during the reform period.

Figure 3

Figure 3. Performance improvement with ITS implementation.

The integration of blockchain technology with ITS also holds significant potential ( Habib et al., 2022 ). Blockchain offers a secure, transparent, and tamper-proof system for recording transactions. When combined with AI, blockchain can ensure that all tax data is recorded in a decentralized ledger that cannot be altered or manipulated. This integration increases the transparency and accountability of tax systems, ensuring that all transactions are traceable and verifiable by auditors ( Eyo-Udo et al., 2025 ). Blockchain’s immutable ledger ensures that data integrity is maintained throughout the tax collection and auditing process, making it harder to manipulate or falsify records.

5. Applications for Intelligent Tax Systems

The implementation of ITS offers numerous applications that transform tax administration. One of the most important applications is automated tax audits ( Saragih et al., 2023 ). Traditional audits are typically resource-intensive, requiring significant human intervention to review financial records and detect discrepancies. However, with ITS, tax authorities can leverage AI to automate this process. AI models can analyze tax returns and compare them against data from other sources, such as bank statements or publicly available financial information ( Cao et al., 2024 ). These systems flag suspicious discrepancies that might otherwise go unnoticed, allowing tax authorities to focus their attention on high-risk cases.

Another significant application is in fraud detection. AI-powered fraud detection systems are capable of identifying suspicious transactions and behaviors that indicate tax evasion or fraud ( Johora et al., 2024 ). By continuously analyzing financial data, ITS can detect anomalies, such as unusually high deductions, discrepancies in reported income, or inconsistencies in tax filings. This proactive approach enables tax authorities to identify fraud earlier and take corrective action before significant revenue losses occur ( Svetlozarova Nikolova, 2023 ). Additionally, AI-based fraud detection systems can adapt over time, learning from new data to improve their accuracy and detection capabilities.

In revenue efficiency, ITS helps improve the allocation of resources by identifying areas where tax collection is lagging. By focusing on high-risk individuals or organizations, tax authorities can ensure that their limited resources are used where they will be most effective ( Chooi, 2020 ). Additionally, ITS can predict future tax revenues, helping governments plan their budgets more accurately. These predictions are based on historical data and trends, making them more reliable than traditional methods.

Finally, ITS can improve taxpayer compliance by simplifying the tax filing process. Many taxpayers struggle to understand their tax obligations, leading to errors or non-compliance. Intelligent systems can guide taxpayers through the filing process, provide reminders about deadlines, and offer real-time assistance in correcting mistakes. By reducing the complexity of tax filing and making it more user-friendly, ITS can enhance voluntary compliance and reduce errors.

6. Challenges in Implementing Intelligent Tax Systems

Despite the promising benefits of ITS, the implementation of these systems presents several significant challenges that need to be carefully addressed. One of the primary challenges is data privacy. Tax data is highly sensitive and includes personal and financial information about individuals and businesses ( Cockfield, 2019 ). The adoption of ITS, which requires large amounts of taxpayer data for analysis and decision-making, raises concerns about the security and confidentiality of this information. Governments must ensure that AI-driven systems comply with data privacy regulations, such as the GDPR in Europe and the CCPA in the United States. Protecting taxpayer data from unauthorized access or misuse is critical for maintaining trust in the system and avoiding legal challenges.

Table 2 summarizes the primary challenges encountered during ITS implementation, categorized by impact level and accompanied by practical solutions. Data privacy and system integration emerge as the most critical challenges, requiring comprehensive planning and substantial time investment to address effectively.

Table 2

Table 2. Major implementation challenges and solutions.

Challenge	Impact Level	Primary Solution	Timeline
Data Privacy	High	GDPR Compliance Framework	6 - 12 months
System Integration	High	Phased Migration Approach	12 - 24 months
Algorithm Bias	Medium	Diverse Training Datasets	3 - 6 months
Staff Training	Medium	Comprehensive Education Programs	6 - 9 months
Cost Constraints	Medium	Public-Private Partnerships	18 - 36 months

Additionally, the integration of ITS with existing tax systems is another major challenge. Many tax authorities still rely on legacy systems that were not designed to work with modern AI technologies ( Ruiz, 2021 ). Upgrading these systems to integrate with new ITS platforms can be costly, time-consuming, and technically complex. The integration process often involves standardizing data formats, ensuring compatibility between different software systems, and retraining staff to work with new technologies. This transition requires significant investment in both technology and human resources, and it may face resistance from tax professionals who are used to traditional methods. Therefore, successful implementation of ITS depends not only on the technology itself but also on the willingness of tax authorities to adapt to a new way of working.

Another challenge is ensuring the fairness and transparency of AI models used in tax administration ( Aladebumoye, 2025 ). Machine learning models are trained on large datasets, and if these datasets contain biases—whether due to historical data or data that is not representative of all demographics—the models can perpetuate these biases in their decisions. For example, an AI system that is used to identify high-risk taxpayers may inadvertently target certain groups more frequently due to biased historical data. This could result in discriminatory practices in areas like audit selection or fraud detection, potentially leading to unfair treatment of taxpayers based on their demographics or socioeconomic status. To prevent such issues, it is essential that tax authorities carefully monitor and audit AI systems to ensure they are fair and non-discriminatory ( Bentley, 2020 ).

Ethical concerns related to the use of AI in tax administration are also an important consideration ( Bishop, 2021 ). AI models, particularly those that rely on predictive analytics, could inadvertently make decisions that are difficult for taxpayers to challenge or understand. For example, if an AI model flags a taxpayer’s return as suspicious, the taxpayer may not have a clear understanding of why the decision was made. Without a transparent explanation of the model’s decision-making process, taxpayers may feel unfairly penalized, leading to mistrust of the system. Ensuring that AI systems provide explainable and transparent decisions is crucial for maintaining taxpayer confidence and ensuring that the system operates in a fair and ethical manner ( Yordanova, 2023 ).

Finally, cost and resource constraints can limit the widespread adoption of ITS. Implementing AI-driven systems requires significant investment in technology, infrastructure, and training. For many governments, especially those in developing countries, the costs associated with adopting ITS may be prohibitive. While the long-term benefits of ITS—such as increased revenue efficiency and improved fraud detection—are clear, the initial investment required to build and implement these systems can be a significant barrier. In addition, many tax authorities may lack the technical expertise needed to develop and deploy these advanced technologies, making collaboration with external vendors or technology partners essential. As such, the successful implementation of ITS requires not only financial investment but also the development of technical capacity within tax authorities.

7. Conclusion

This study acknowledges several methodological limitations that provide opportunities for future research. First, the narrative review approach, while comprehensive, may not capture all relevant studies due to publication bias toward successful implementations. Future systematic reviews with meta-analytical approaches could provide a more robust quantitative synthesis of ITS effectiveness across different contexts.

Second, the empirical evidence presented relies heavily on case studies and government reports, which may suffer from selection bias and limited external validity. Longitudinal studies with control groups are needed to establish causal relationships between ITS implementation and revenue outcomes while controlling for confounding factors such as economic cycles and policy changes.

Third, the lack of standardized metrics across jurisdictions limits comparative analysis. Future research should develop standardized performance indicators for ITS evaluation, enabling more rigorous cross-country comparisons and evidence-based policy recommendations.

Fourth, this study focuses primarily on technical and operational aspects while giving limited attention to behavioral and organizational factors affecting implementation success. Future research should examine how institutional capacity, staff training, and change management practices influence ITS adoption outcomes.

Finally, the rapid evolution of AI technologies means that current findings may become outdated quickly. Continuous monitoring studies and real-time evaluation frameworks are needed to track the evolving landscape of AI applications in tax administration and their long-term societal impacts.

In conclusion, ITS offers transformative benefits for the tax administration process, enhancing the accuracy, speed, and efficiency of tax audits, improving fraud detection, and optimizing revenue collection. The ability of ITS to automate complex processes, analyze large datasets, and predict tax evasion has the potential to significantly reduce the burden on tax authorities while improving compliance and transparency in tax systems. Moreover, by automating routine tasks and focusing resources on high-risk cases, ITS can maximize the efficiency of tax collection and ensure that governments can better allocate their resources.

However, the implementation of ITS is not without its challenges. Data privacy, system integration, bias in AI models, and costs associated with adopting new technologies all pose significant barriers to successful deployment. Additionally, ethical concerns related to transparency and fairness must be carefully managed to ensure that AI systems do not perpetuate discriminatory practices. As governments continue to explore and implement ITS, these challenges will need to be addressed to fully realize the potential of AI in tax administration.

Despite these hurdles, the future of Intelligent Tax Systems looks promising.

The future of intelligent tax administration lies not in replacing human judgment with artificial intelligence, but in creating synergistic human-AI systems that combine computational efficiency with ethical oversight, transparency, and accountability. Success will depend on governments’ ability to balance technological innovation with fundamental principles of fairness, privacy, and taxpayer rights.

As technology continues to evolve, the capabilities of ITS will expand, potentially integrating with blockchain technology and big data analytics to create even more transparent, secure, and efficient tax systems. Governments must prioritize investment in these technologies while also focusing on regulatory compliance, data security, and ethical AI practices. By overcoming these challenges, ITS can become a cornerstone of modern tax administration, contributing to fairer, more efficient, and more trustworthy tax systems around the world.

References

1. Aladebumoye, T. (2025). The Role of AI in Enhancing Tax Transparency and Reducing Evasion. World Journal of Advanced Research and Reviews, 25, 206-212.

2. Ariyibi, K. O., Bello, O. F., Ekundayo, T. F.,&Ishola, O. (2024). Leveraging Artificial Intelligence for Enhanced Tax Fraud Detection in Modern Fiscal Systems. GSC Advanced Research and Reviews, 21, 129-137.

3. Belahouaoui, R.,&Attak, E. H. (2024). Digital Taxation, Artificial Intelligence and Tax Administration 3.0: Improving Tax Compliance Behavior—A Systematic Literature Review Using Textometry (2016-2023). Accounting Research Journal, 37, 172-191. >https://doi.org/10.1108/arj-12-2023-0372

4. Bello, O. A., Folorunso, A., Onwuchekwa, J., Ejiofor, O. E., Budale, F. Z.,&Egwuonwu, M. N. (2023a). Analysing the Impact of Advanced Analytics on Fraud Detection: A Machine Learning Perspective. European Journal of Computer Science and Information Technology, 11, 103-126.

5. Bello, O. A., Ogundipe, A., Mohammed, D., Adebola, F.,&Alonge, O. A. (2023b). AI-Driven Approaches for Real-Time Fraud Detection in US Financial Transactions: Challenges and Opportunities. European Journal of Computer Science and Information Technology, 11, 84-102.

6. Bentley, D. (2020). Taxpayer Rights and Protections in a Digital Global Environment. In R. van Brederode (Ed.), Ethics and Taxation (pp. 251-294). Springer Singapore. >https://doi.org/10.1007/978-981-15-0089-3_11

7. Bezditnyi, V. (2024). Use of Artificial Intelligence for Tax Planning Optimization and Regulatory Compliance. Research Corridor Journal of Engineering Science, 1, 103-142.

8. Bishop, E. (2021). Legal Issues Arising from the Use of Artificial Intelligence in Government Tax Administration and Decision Making. Journal of AI, Robotics&Workplace Automation, 1, 99-108. >https://doi.org/10.69554/ejlj8698

9. Cahyadi, R. (2024). Taxes and Infrastructure Development: A Review from Various Countries. Advances in Taxation Research, 2, 117-127. >https://doi.org/10.60079/atr.v2i3.349

10. Cao, S. S., Jiang, W., Lei, L.,&Zhou, Q. (2024). Applied AI for Finance and Accounting: Alternative Data and Opportunities. Pacific-Basin Finance Journal, 84, Article ID: 102307. >https://doi.org/10.1016/j.pacfin.2024.102307

11. Çetin Gerger, G. (2019). Tax Services and Tax Service Providers’ Changing Role in the IoT and Ami Environment. In Z. Mahmood (Eds.), Guide to Ambient Intelligence in the IoT Environment (pp. 203-216). Springer International Publishing. >https://doi.org/10.1007/978-3-030-04173-1_9

12. Chooi, A. (2020). Improving Tax Compliance: Establishing a Risk Management Framework. The Governance Brief (No. 39).

13. Cockfield, A. J. (2019). Sharing Tax Information in the 21st Century: Big Data Flows and Taxpayers as Data Subjects. Canadian Tax Journal/Revue Fiscale Canadienne, 67, 1179-1199. >https://doi.org/10.32721/ctj.2019.67.4.sym.cockfield

14. Đukić, T., Pavlović, M.,&Grdinić, V. (2023). Uncovering Financial Fraud: The Vital Role of Forensic Accounting and Auditing in Modern Business Practice. Economic Themes, 61, 407-418. >https://doi.org/10.2478/ethemes-2023-0021

15. Eyo-Udo, N. L., Apeh, C. E., Bristol-Alagbariya, B., Udeh, C. A.,&Ewim, C. P. M. (2025). The Evolution of Blockchain Technology in Accounting: A Review of Its Implications for Transparency and Accountability. Account and Financial Management Journal, 10, 3467-3478.

16. Ezeife, E., Kokogho, E., Odio, P. E.,&Adeyanju, M. O. (2021). The Future of Tax Technology in the United States: A Conceptual Framework for AI-Driven Tax Transformation. International Journal of Multidisciplinary Research and Growth Evaluation, 2, 542-551. >https://doi.org/10.54660/.ijmrge.2021.2.1.542-551

17. Ezeife, E., Kokogho, E., Odio, P. E.,&Adeyanju, M. O. (2025). Agile Tax Technology Development in the U.S.: A Conceptual Framework for Scalable and Efficient Enterprise Solutions. Gulf Journal of Advance Business Research, 3, 512-526. >https://doi.org/10.51594/gjabr.v3i2.94

18. Fedyk, A., Hodson, J., Khimich, N.,&Fedyk, T. (2022). Is Artificial Intelligence Improving the Audit Process? Review of Accounting Studies, 27, 938-985. >https://doi.org/10.1007/s11142-022-09697-x

19. Folorunso, B. A.,&Nwankwo, C. B. (2024). Harnessing the Power of AI for Taxation: Nigeria’s Path to Improved Revenue Generation and Transparency. Nnamdi Azikiwe University Journal of Private and Property Law, 1, 113-123.

20. Habib, G., Sharma, S., Ibrahim, S., Ahmad, I., Qureshi, S.,&Ishfaq, M. (2022). Blockchain Technology: Benefits, Challenges, Applications, and Integration of Blockchain Technology with Cloud Computing. Future Internet, 14, Article 341. >https://doi.org/10.3390/fi14110341

21. Hilal, W., Gadsden, S. A.,&Yawney, J. (2022). Financial Fraud: A Review of Anomaly Detection Techniques and Recent Advances. Expert Systems with Applications, 193, Article ID: 116429. >https://doi.org/10.1016/j.eswa.2021.116429

22. Johora, F. T., Hasan, R., Farabi, S. F., Akter, J.,&Mahmud, M. A. A. (2024). AI-Powered Fraud Detection in Banking: Safeguarding Financial Transactions. The American Journal of Management and Economics Innovations, 6, 8-22. >https://doi.org/10.37547/tajmei/volume06issue06-02

23. Kirer, G. (2024). Agency Revenue Collection, Automation and Own Source Revenue Targets among Selected County Governments in Kenya. Master’s Thesis, UOK.

24. Nurbekova, G. T., Juchnevicius, E. Е.,&Tussupova, L. K. (2024). An Integrated Approach to the Concept and Elements of Tax Administration: Theoretical and Legal Aspects. Bulletin of the Karaganda University. “Law Series”, 114, 16-24.

25. Okunogbe, O.,&Santoro, F. (2023). Increasing Tax Collection in African Countries: The Role of Information Technology. Journal of African Economies, 32, i57-i83. >https://doi.org/10.1093/jae/ejac036

26. Olabanji, S. O., Olaniyi, O. O.,&Olaoye, O. O. (2024). Transforming Tax Compliance with Machine Learning: Reducing Fraud and Enhancing Revenue Collection. SSRN Electronic Journal.

27. Rahman, S., Sirazy, M. R. M., Das, R.,&Khan, R. S. (2024). An Exploration of Artificial Intelligence Techniques for Optimizing Tax Compliance, Fraud Detection, and Revenue Collection in Modern Tax Administrations. International Journal of Business Intelligence and Big Data Analytics, 7, 56-80.

28. Ruan, J., Yan, Z., Dong, B., Zheng, Q.,&Qian, B. (2019). Identifying Suspicious Groups of Affiliated-Transaction-Based Tax Evasion in Big Data. Information Sciences, 477, 508-532. >https://doi.org/10.1016/j.ins.2018.11.008

29. Ruiz, M. A. G. (2021). Fiscal Transformations Due to AI and Robotization: Where Do Recent Changes in Tax Administrations, Procedures and Legal Systems Lead Us? Northwestern Journal of Technology and Intellectual Property, 19, Article 1.

30. Saragih, A. H., Reyhani, Q., Setyowati, M. S.,&Hendrawan, A. (2023). The Potential of an Artificial Intelligence (AI) Application for the Tax Administration System’s Modernization: The Case of Indonesia. Artificial Intelligence and Law, 31, 491-514. >https://doi.org/10.1007/s10506-022-09321-y

31. Scientific, L. L. (2024). Optimization of Smart Taxation Using Artificial Intelligence: Risks and Opportunities. Journal of Theoretical and Applied Information Technology, 102, 1870-1884.

32. Svetlozarova Nikolova, B. (2023). Improving Prevention and Diagnostics in Tax Control. In B. Svetlozarova Nikolova (Ed.), Tax Audit and Taxation in the Paradigm of Sustainable Development (pp. 119-136). Springer Nature Switzerland. >https://doi.org/10.1007/978-3-031-32126-9_6

33. Tomar, M.,&Periyasamy, V. (2023). The Role of Reference Data in Financial Data Analysis: Challenges and Opportunities. Journal of Knowledge Learning and Science Technology, 1, 90-99. >https://doi.org/10.60087/jklst.vol1.n1.p99

34. Yordanova, Z. (2023). Ethical Implications of Transparency and Explainability of Artificial Intelligence for Managing Value-Added Tax (VAT) in Corporations. In T. Guarda, F. Portela,&J. M. Diaz-Nafria (Eds.), Advanced Research in Technologies, Information, Innovation and Sustainability. ARTIIS 2023 (pp. 344-353). Springer Nature Switzerland. >https://doi.org/10.1007/978-3-031-48855-9_26

35. Zheng, Q., Xu, Y., Liu, H., Shi, B., Wang, J.,&Dong, B. (2024). A Survey of Tax Risk Detection Using Data Mining Techniques. Engineering, 34, 43-59. >https://doi.org/10.1016/j.eng.2023.07.014