Technological Crossroads Of GST - How To Get The Balance Right
Author: LawSphereX
Category: Taxation
Published:
Updated:
TECHNOLOGICAL CROSSROADS OF GST: HOW TO GET
THE BALANCE RIGHT
Akriti Sharma, LL.M. (Taxation Law), Jindal Global Law School, Sonipat
ABSTRACT
A major part of how people live today and carry out their businesses is
dependent on the digital economy. With the rapid growth of the digital economy, the way goods and services are produced, or brought or sold, and how businesses are carried out globally are transforming. As India has one of the largest and fastest-growing digital markets in the world, it is important to understand how technology is playing an important role in shaping the taxation system today. The Goods and Services Tax (GST) system was introduced in India to make the indirect taxation administrative system at par with the changing nature of taxation system in this digital age. The taxation industry is also changing significantly, and machine learning (ML) and artificial intelligence (AI) are becoming more and more crucial in expediting compliance procedures.
The present paper aims to address the legal and technological framework of GST and attempts to highlight a significant gap that exists between the intended goals and objectives of GST and the actual experience of the taxpayers in complying with the GST laws. It also aims to analyze a survey that was conducted in order to understand the perception of the people towards the GST administration system and the integration of AI in the GST framework. Keywords: Technological Framework, Digital Economy, Artificial Intelligence, Goods and Services Tax, Goods and Services Tax Network, Administration
INTRODUCTION
As the digital economy continues to grow, the need for a tech-based administrative framework has become the need of the hour. Technology is being used by countries around the world to modernize their taxation systems in order to enhance compliance and reduce tax evasion. An example of this is the introduction of the Goods and Services Tax (GST) system in India which has made the indirect taxation system at par with the changing nature of taxation system in this digital age. GST was introduced in the year 2017 within the Indirect Taxation system in India, and since its introduction, the landscape of GST has transformed, bringing in more efficiency, transparency and accountability in the administrative system.
With the advancement of technology across the globe, various developments such as the Goods and Services Tax Network (GSTN), e-invoicing, e-way bills, data analytics, integration of AI within the GST framework are playing an important role in revolutionizing the GST administrative system in a digitalized and globalized economy today. In other words, it is important to understand how these technological advancements are shaping the tax administration system today. The GSTN has increased transparency and also reduced the burden of compliance on the taxpayers to a great extent by enhancing transparency and efficiency in tax administration system.
The technological advancements within the GST framework are beyond compliance as it has also introduced various services for the taxpayers in order to facilitate a better communication with the tax authorities and to build the trust of the taxpayers in the administrative system. Although GST has made the tax administrative system more transparent and accountable, it also comes with its own set of challenges, especially with regard to its implementation and compliance mechanisms. The present paper addresses the legal and technological framework of GST attempts to highlight a significant gap that exists between the intended goals and objectives of GST and the actual experience of the taxpayers in complying with the GST laws. For this, the paper consists of four chapters. The first chapter aims to address the legal and technological
framework of GST, followed by the second chapter which deals with the integration of AI within the framework of GST. The third chapter deals with the question of equitable distribution of benefit between the taxpayers and the tax authorities. And, lastly, the fourth chapter aims to analyze the survey that was conducted in order to understand the perception of the people towards the GST administration system and the integration of AI in the GST framework.
CHAPTER 1: LEGAL & TECHNOLOGICAL
FRAMEWORK OF GST
The present chapter aims to provide the basis for understanding the current technological and legal framework of GST and serves as the theoretical foundation for a more detailed analysis of the issues outlined in the subsequent chapters. A major part of how people live today and carry out their businesses is dependent on the digital
economy, which includes e-commerce, cloud computing, online services and digital platforms, use of big data, etc. Today, the digital economy is considered as an economy in itself. As the digital economy is rapidly growing, it is also transforming the way goods and services are produced, or brought or sold, and how businesses are carried out globally. Technology is omnipresent and has entered every sphere of commerce and business today as from developed to developing economies, technology and innovation are redefining the various stages of production which are not only offering cost-effective solutions, but also fostering new possibilities in terms of communication and business communication.
1 The global landscape of taxation has witnessed a lot of changes in its administration system, especially because tax administrations in the whole world has been continuously trying to balance their system with that of the technological advancements, in order to enhance the effectiveness and accountability of their administrative system. 2 India’s tax administration has undergone significant evolution since independence, especially with the formation of Central Board of Revenue in 1944, which later split into the CBDT and 1 Krishan Arora et al, GST and Technology: All you need to know, [2022] 138 taxmann. com 551 (Article), 30 May 2022. 2 CA Mayank Mohanka, Digital Transformation of Indian Tax Administration: CBDT’s ‘Faceless e Assessments’ & CBIC’s e-Invoicing, [2019] 112 taxmann. com 234 (Article), 17 December 2019.
CBIC in 1964. 3 This has established the foundational structure of the Indian tax administration today as for decades, tax administration remained largely paper-based, characterized by manual processes, limited enforcement capabilities, and significant compliance gaps. 4 As India has one of the largest and fastest-growing digital markets in the world, it is important to understand how technology is playing an important role in shaping the taxation system today. At the same time, technology is being used by countries around the world to modernize their taxation systems in order to enhance the compliance requirements and reduce the evasion of tax. The Goods and Services Tax (GST) system was introduced in India within the Indirect Taxation system in order to make the entire administrative system at par with the changing nature of taxation system in this digital age.
The Goods and Services Tax (GST) is a landmark reform in indirect taxation in India and it has been framed in such a manner that its implementation is heavily reliant on technology. 5 GST is an indirect tax on the supply of goods and services within India. 6 The Central Goods and Services Tax Act was passed on 29th March 2017 which came into force on 1st July 2017. 7 GST subsumed 17 different indirect taxes and introduced uniformity in the indirect tax system throughout the country by removing the cascading effect of tax, i. e. “tax on tax”. 8 Thus, the primary aim behind introducing GST was to simplify the tax structure, eliminate cascading effects, promote ease of doing business, and create a unified market across the country.
9 Prior to the implementation of GST in the country, there was no common interface under the indirect tax regime where the tax authorities as well as the taxpayers could access all the tax services. The introduction of GSTN i. e. Goods and Services Tax Network, within the GST framework has brought the taxpayers and the tax authorities under the same roof by providing a common interface where everyone is able to access the GST services at the same time. The current framework of GST comprises of the GSTN, which is the foundational stone and the
3 Section 3, The Central Board of Revenue Act, 1963, Act No. 54 of 1963
4 CS Charu Vinayak, Artificial Intelligence and Tax Compliance, Chartered Secretary, April 2025.
5 CS Charu Vinayak, Artificial Intelligence and Tax Compliance, Chartered Secretary, April 2025.
6 Annapoorna, Goods and Services Tax: What is GST in India? Indirect Tax Law Explained, ClearTax, (May 4, 2025, 8:30 PM), https://cleartax.in/s/gst-law-goods-and-services-tax.
7 The Central Goods and Services Tax Act, 2017, Act No. 12 of 2017.
8 Cleartax, Simplification of Indirect Taxes under GST Regime, (Jan. 23, 2025), https://cleartax.in/s/cascadingeffect-under-current-indirect-tax-regime. 9 Dr. B Ramaswamy, GST: New Legal Initiatives and its Implications, [2024] 158 taxmann.com 709 (Article), (Feb. 1, 2024).
technological backbone for streamlining the indirect tax system in the country. The GST Common Portal under the GSTN serves as the front end of the whole GST IT eco-system. 10 The GSTN has been set up as a Section 25 not-for-profit company, in which the Government of India and State Government hold 49% of the shareholding, and the balance 51% is held by corporations and banks such as LIC, ICICI, HDFC etc. 11 The GSTN acts as the technological backend of the GST administration system and the contract for developing it was given to the IT company, Infosys in the year 2015. 12 The most important objective of introducing GSTN within the indirect tax system was to provide a common interface for filing of returns, e-invoicing, payment of penalties and interest, filing of appeals against the orders and notices passed by the tax authorities, etc.
Thus, the GSTN is the comprehensive back-end infrastructure for the management of data related to tax and reporting of GST. 13 The GSTN portal not just acts as a technology-based solution provider but also as a ‘catalyst’ for making the tax administrative system more efficient and transparent. 14 Thus, the it plays a ‘multifaceted role’ by acting as a doorway for complying with the GST laws in India. 15 The GSTN also serves as a “trusted National Information Utility” which provides an efficient as well as a reliable framework for the proper and smooth functioning of the GST system in India. 16 One of the primary functions of the GSTN is create an efficient as well as user-friendly GST eco-system and at the same time provide support to the tax authorities in order to bring in transparency in the tax administration system and also enhance the compliance mechanisms. 17
10 Pavan Kumar. R, GST Administration Through the Goods and Services Tax Network: A Critical
Analysis, [2021] 127 taxmann.com 759 (Article), (Jun. 21, 2021). 11 Karthik Sundaram, GSTN- The New Network, National Law School of India Review, Vol. 28, No. 2 (2016), pp. 114-122. 12 ClearTax, What is GSTN (Goods and Services Tax Network)?,(Jan. 08, 2025), https://cleartax.in/s/what-is-gstn. 13 Dr. Sanjiv Agarwal, GSTN Common Portal – Salient features (Part-I), Tax Management India.com, (Oct. 27, 2017). 14 ClearTax, What is GSTN (Goods and Services Tax Network)?,(Jan. 08, 2025), https://cleartax.in/s/what-is-gstn. 15 Abhinandan Banerjee, GSTN: The Technological Backbone of India’s GST System-A Comprehensive Guide, (Feb. 23, 2025), https://authbridge.com/knowledge-centre/glossary/gst-council-meetings. 16 ClearTax, What is GSTN (Goods and Services Tax Network)?,(Jan. 08, 2025), https://cleartax.in/s/what-is-gstn.
17 Pavan Kumar. R, GST Administration Through the Goods and Services Tax Network: A Critical
Analysis, [2021] 127 taxmann.com 759 (Article), (Jun. 21, 2021).
One of the projects of the GSTN in order to improve compliance and minimize tax avoidance is the ‘e-invoicing’ system, which was introduced as per the recommendations of the 37th GST Council Meeting on September 20th, 2019. 18 In addition to this, the introduction of a separate e-invoice portal has helped in minimizing fraud and generation of bogus invoices as every invoice is being verified and reported to the to the Invoice Registration Portal (IRP) which further issues a unique Invoice Reference Number (IRN) and a digitally signed QR code. 19 Another initiative of the GSTN is the introduction of ‘e-way bill’ which has to be mandatorily generated before the transportation of goods from one place to another and acts a document evidencing the movement of the goods.
It is an electronic document which is being generated on a separate e-way bill portal and ensures that all the information related to the goods such as the GSTN number, value of goods, invoice number, HSN Code etc, are authentic in order to keep a track on the movement of goods and reduce the evasion of tax. The technological framework of GST has also made it easier for the tax authorities to scrutinize the e-way bills, e-invoices, matching of GSTR-2A and GSTR -3B, etc. GSTN is also the first IT related development in the indirect tax regime that aims to provide a consistent interface for the taxpayer as well as a single IT infrastructure that is shared between the State and the Centre.
20 Therefore, with the implementation of GST and its overreaching goal of reducing the multiplicity of taxes, the GSTN is tasked with the difficult task of integrating the various and disparate systems of the Centre and the States, as well as bringing all the tax administrations at an equal level of technological maturity that has the same interface for the taxpayers and other stakeholders. 21 Another important project of the GSTN is the Tax Information Exchange System (TINXSYS), which acts a centralized system for smooth administration of trade between the states and provides as a framework for ensuring effective exchange of information among all the states the Commercial Tax Departments (CTDs). 22 The Tax Information Exchange System is a
18 Goods and Services Tax Network (A Government Enterprise), https://www.gstn.org.in/.
19 Goods and Service Tax, E-invoice System, https://einvoice.gst.gov.in/einvoice/overview.
20 About Us, Genesis and Structure of GSTN, https://www.gstn.org.in/about-us.
21 About Us, Genesis and Structure of GSTN, https://www.gstn.org.in/about-us.
22 Goods and Services Tax Network (A Government Enterprise), https://gstn.org.in/.
technology-based platform that was established in the year 2004 by the Empowered Committee of State Finance Ministers for gathering information and facilitating inter-state trade. 23 Another pillar of GST, which also acts as the heart of the GST framework is the Input Tax Credit (ITC) System. It is purely a technology-based compliance mechanism which allows the registered businesses or persons to claim credit for the taxes paid on the inputs used or intended to be used in the course of furtherance of business. 24 The introduction of an automated system that matches the invoices uploaded and GSTR 1 filed by the supplier with that of the returns filed by the recipient in GSTR-3B and creation of an auto generated statement of eligible ITC in GSTR-2B, has helped in enhancing the compliance requirements and encouraging the businesses to get involved only in genuine transactions.
Furthermore, the GSTN has also issued an advisory for initiating a ‘case-insensitive’ einvoicing system in order to generate the Invoice Reference Number (IRN) through the Invoice Registration Portal (IRP), which is supposed to take effect from 1st June 2025. 25 This change in the e-invoicing system is mainly introduced to minimize the discrepancies caused in the IRNs because of character casing, which will make the generating of invoices more aligned with the GSTR-1 filing.
26 Therefore, the main objective of the GST system was to bring ease and transparency in the indirect tax regime and in this regard the online platform was developed within the GST system which is accessible through laptops, mobile phones and it provides a number of offline tools for filing the return within the due date and has made all of its return and ledger and features available to various software via the Application Programming Interface (API). And with the ease of online filing of returns, appeals to the orders and notices, and simplifying the entire multi-layered dispute resolution process that prevailed in the old regime, the GSTN has played a vital role in modernising the tax administration and dispute resolution mechanism within the indirect tax system. It has not only increased transparency but also reduced the burden of compliance on the taxpayers to a great extent.
It has also reduced paperwork and
23 Goods and Services Tax Network (A Government Enterprise), https://gstn.org.in/.
24 Pramit Pratim Ghosh, Input Tax Credit (ITC) under GST, (Aug. 11, 2023), https://tallysolutions.com/gst/gstinput-tax-credit/?srsltid=AfmBOoryWa6GVbu_wex7wIll2TUxClEJda1XkEmdeULJQKttA7RLx5_W. 25 Renu Suresh, GSTN Introduces Case-Insensitive E-Invoicing System Effective June 1, 2025, India Filings, (Apr 08, 2025), https://www.indiafilings.com/learn/case-sensitive-gst-einvoicing-system/. 26 Renu Suresh, GSTN Introduces Case-Insensitive E-Invoicing System Effective June 1, 2025, India Filings, (Apr 08, 2025), https://www.indiafilings.com/learn/case-sensitive-gst-einvoicing-system/.
human intervention, thereby bringing in more transparency and efficiency in tax administration. Moreover, there are GST Suvidha Providers (GSP) as well as the Application Service Providers (ASP) which provide assistance to the taxpayers for the compliance matters. And, in order to ensure that the taxpayer’s data is kept anonymous, the APIs are designed in such a way that it can take encrypted data from Application Service Providers or the GST Suvidha Providers, which are decrypted only by the GST System. For instance, a unique framework for API is designed, which has been deployed in various GSP and Government systems. 27 Through the Application Software Platform (ASP) solution, the GSPs are aimed to provide an interface through which the businesses will be able to communicate with the GSTN.
28 The GSTN has also introduced a new functionality called the Invoice Management System (IMS), effective from 1st October 2024, which aims to revolutionize how businesses handle invoices and claim Input Tax Credit (ITC) under the GST regime. 29 The IMS aims to provide a platform where the taxpayers can easily match and verify the invoices raised by the suppliers before they form a part of the ITC claims. Therefore, the taxpayers are able to handle the invoice related activities through the IMS. This development within the GSTN framework has played a very important role in providing an opportunity for the taxpayers to match and verify their invoices in order to prevent any future discrepancies related to mismatch of GSTR-2A and 3B.
Furthermore, the government introduced the IMS as one of the developments within the GST framework that provides for a platform wherein the suppliers and the recipient are able to resolve any issues with regard to the invoices and thereby facilitating better communication within the system. 30 Below is the statistics of the GST system as on 31st March 2025, provided by the Goods and
27 GSTN- Goods and Service Tax Network, GST System Architecture and Rollout Strategy,
https://www.gstn.org.in/assets/mainDashboard/Pdf/GST%20Technology%20Architecture.pdf.
28 Amit Sarkar, How Technology Solutions Can Help Businesses Create Their GST Framework, BW
Businessworld, (Aug. 22, 2017), https://www.businessworld.in/article/how-technology-solutions-can-helpbusinesses-create-their-gst-framework-124472.
29 CA Surjeet Singh Negi, Invoice Management System- A paradigm shift in GST Compliances, [2024] 166
taxmann.com 423 (Article), (Sep. 18, 2024).
30 Karan Kakkar and Pranav Aggarwal, Invoice Management System: Unlocking efficiency for
regular GST compliances? , [2024] 167 taxmann.com 275 (Article), (Oct. 10, 2024).
Services Tax Network31: It is not just the GSTN that is functioning alone to enhance the compliance mechanism and bring in transparency and accountability within the indirect tax system. The states also equally play an important role in carrying out other statutory functions. But, considering the diversity that exists amongst the taxpayers in our country because of the different kinds of businesses that are being performed, it was never easy to implement a technology-based taxation system that could equally benefit everyone. This vast range of differences amongst the taxpayers has created a lot of challenges, especially during the initial stages of implementation of GST as the small businesses were facing difficulties with the sudden transition from manual operations to automated accounting systems.
Therefore, it is important to understand how technological advancements such as the Goods and Services Tax Network (GSTN), e-invoicing, e-way bill, Input Tax Credit (ITC) or the integration of Artificial Intelligence (AI) is shaping the tax administration system today. These developments within the indirect taxation system are playing a significant role in providing a common interface which is accessible to all the taxpayers in different parts of the country at the same time, and also facilitating easy communication between the taxpayers and the tax authorities. But every technological development comes with its own set of challenges. The next chapter aims to understand and analyse how Artificial Intelligence (AI) is transforming the present
31 Goods and Services Tax Network (A Government Enterprise), https://gstn.org.in/.
administrative system, and creating concerns amongst the taxpayers about the confidentiality of their data and information and also the biases that may be created by the AI algorithms.
CHAPTER 2: THE INTEGRATION OF AI IN GST ADMINISTRATION
The present chapter aims to understand the use of Artificial Intelligence (AI) within the GST framework and how it is shaping the tax administrative system today. The emergence of Artificial Intelligence (AI) has significantly transformed various facets of human life.32 John McCarthy was the first person to coin the term ‘artificial intelligence’ in
195633 at the Dartmouth Conference by defining it as “the science and engineering of making
intelligent machines which encompasses the development of computer systems capable of tasks that typically require human intelligence, such as visual perception, speech recognition, decision- making, and language translation. 34 Thus, AI refers to the utilization of technologies like machine learning to stimulate human intelligence and decision-making processes. 35 AI adoption in tax compliance is not limited to India. Countries like the US, Canada and the UK have already integrated AI-driven solutions into their tax regimes. 36 If we combine AI with the “Internet of Things (IoT)”, “Data Analysis and Data Analytics” and use it on the information received from the taxpayers, it helps in creating various advantages in the administration processes and provides assistance in resolving various administrative disputes.
37 The taxation industry is changing significantly, and machine learning (ML) and artificial intelligence (AI) are becoming more and more crucial in expediting compliance procedures. The present chapter, therefore, talks about the integration of AI within the GST framework and how biases in the AI system might affect the taxpayers and the economy as a whole.
In India, where taxation laws are intricate and ever-evolving, AI has the potential to revolutionize tax management for businesses and professionals. 38 Today, AI has brought in revolutionary solutions to various complexities arising in the GST framework. AI can filter through huge volumes of transactional data by using advanced algorithms and machine learning models to identify and detect anomalies or any fraudulent activity, and at the same time also predict if there is any potential risk in the compliance mechanism. There have been instances wherein the tax administrations today are adopting various advanced analytics techniques and data mining processes in order to enable seamless collection of revenue and are able to collect data from the businesses on a real-time basis without any direct connection with the taxpayers.
39 The GST system generates vast amounts of transactional data, creating opportunities for AI driven analytics40. The integration of AI into the GST administration is one of the important technological developments within the indirect taxation system aimed at enhancing the administrative system and compliance mechanisms. Research by the National Institute of Public Finance and Policy indicates that AI-driven analytics in GST administration has contributed to a 12% increase in compliance rates among previously high-risk sectors41. The GSTBOT which is a virtual assistant for GST-related inquiries is capable of answering questions about filing requirements or any payment related procedures or policy clarifications in multiple Indian languages, and is an AI-powered virtual assistance which have been deployed to improve the taxpayers’ services.
42 The integration of AI within the GST framework has made the administration system more transparent and accountable as it has not only enhanced the efficiency of the administration of GST but has also strengthen the integrity of the GST system by ensuring that the administrative burden is reduced on the taxpayers and the tax obligations are properly carried out by them. AI also plays a vital role in the fights against tax evasion through the detection of fraudulent activities, real-time monitoring, cross-verification of data, risk assessment and profiling, 38 CS Suparn Sekhri, Artificial Intelligence and Tax Compliance: A Paradigm Shift, Chartered Secretary, April 2025. 39 CA Mayank Mohanka, Digital Transformation of Indian Tax Administration: CBDT’s ‘Faceless e Assessments’ & CBIC’s e-Invoicing, [2019] 112 taxmann. com 234 (Article), (Dec. 17, 2019).
40 CS Charu Vinayak, Artificial Intelligence and Tax Compliance, Chartered Secretary, April 2025.
41 CS Charu Vinayak, Artificial Intelligence and Tax Compliance, Chartered Secretary, April 2025.
42 CS Charu Vinayak, Artificial Intelligence and Tax Compliance, Chartered Secretary, April 2025.
automation of audits and in many other ways. 43 For instance, Maharashtra government was the first to use AI and other technological tools to trace GST frauds. 44 Similarly, the multi-national corporations all over the world deal with huge amount of financial data on an everyday basis, and AI-powered data analytics tools has helped these corporation to a great extent by making the compliance mechanisms more easily understandable and useable and also fulfilling their tax obligations without creating any complexities. It can also be said that machine learning (ML), when being employed in order to detect any fraud has the capacity to analyse huge data and can easily identify any fraudulent activity which might not have been easily detectable under any conventional method.
45 Furthermore, the AI-based tools and softwares have the ability to provide assistance to the businesses for streamlining their compliance requirements with ease. 46 Another advantage of integrating Generative AI within the GST framework is the that it has helped in reducing time and effort for reconciliation of data to a great extent and at the same time identifying discrepancies in the invoices, thereby reducing the likelihood of erroneous claims. 47 Thus, from collection of data to filing of returns, these AI tools are helping the taxpayers at every stage of the compliance process. With AI influencing almost every aspect of our life today, it can be said that the contribution of AI towards tax compliance extends beyond simple automation.
It is not only helping the taxpayers in identifying any irregularities in the tax returns and avoiding any interests and penalties, but also working as an aid to the tax authorities in identifying any fraudulent activity very easily. But, the recent flooding of notices and demand orders across different industries in India has raised several concerns in the use of AI and data analytics by the tax authorities. Tax demands 43 Dr. Keerthy V. S. et al, Artificial Intelligence Law & Ethics, MPP House, First Edition 2024, pg- 260-261. 44The Times of India, Maharashtra using AI tools to detect GST frauds : CM, (Dec. 21, 2024), https://timesofindia. indiatimes. com/city/nagpur/maharashtra-using-ai-to-detect-gst-fraudscm/articleshow/116514635. cms.
45 Anju Sharma, Dr. Kapil Kumar, Detecting GST Fraud Through Machine Learning Techniques, Journal of
Emerging Technologies and Innovative Research (JETIR), 2021 JETIR March 2021, Volume 8, Issue 3. 46 RMPS & Co. Simplifying GST Reconciliation and Return Filing with AI: A Guide for Businesses, (May 3, 2024), https://rmpsco.com/simplifying-gst-reconciliation-and-return-filing-with-ai-a-guide-for-businesses/. 47 Abhinit Singh, Becoming a Future-Ready Accountant: How Generative AI is Integrating into GST Management in India, (Sep 15, 2023), https://www.telegraphindia.com/edugraph/career/becoming-a-future-ready-accountanthow-generative-ai-is-integrating-into-gst-management-in-india/cid/1966404.
from industries like online gaming, shipping, and airlines have increased as a result of the government's greatly increased scrutiny of tax records. Thus, the use of AI to evaluate enormous volumes of financial data is one of the main causes of this surge in the GST notices. The way the government is collecting GST is significantly changing as AI is integrating more and more into the GST framework. We can also take the example of an initiative taken by the Uttar Pradesh Government wherein in the FY 2023–2024, the AI module that they created contributed to the generation of approximately 6 lakhs notices and as a result of this, the department received around ₹980 crore as tax deposits. 48 The Additional Chief Secretary, Housing and Urban Planning and State Tax Department of the Government of Uttar Pradesh, Mr.
Nitin Ramesh Gokarn, in one of his interviews talked about how there was a 21% growth in the collection of revenue by the Uttar Pradesh state government in FY 2023-24, and how Artificial Intelligence was deployed in the systems which led to the increase in the collection of GST. 49 He further stated that “taking a cue from the best practices of Telangana and Andhra Pradesh, where they had developed an Artificial Intelligence tool which helps in data mining and generating automated notices, Uttar Pradesh also decided to incorporate the model within its system and accordingly, the data mining from the GST data was done through the AI tool. ”50 It can therefore be said that the government’s efficiency, accuracy, and revenue collection have increased as a result of the application of AI in GST administration.
Even though AI clearly benefits the government, it also raises questions about transparency and fairness as the AI algorithms can be easily manipulated to create bias. Therefore, it can be said that, although the AI-driven solutions have the capability to improve the indirect taxation system, it also creates serious concerns as the issue of bias in the AI system can undermine the GST administration system. 48 Arpit Gupta, GST: Deploying AI solutions for data mining, generating automated notices & driving government revenues, ETGovernment, (May 24, 2024), https://government. economictimes. indiatimes. com/news/technology/gst-deploying-ai-solutions-for-data-mininggenerating-automated-notices-driving-govt-revenues/110370822.
49 Arpit Gupta, GST: Deploying AI solutions for data mining, generating automated notices & driving government revenues, ETGovernment, (May 24, 2024), https://government. economictimes. indiatimes. com/news/technology/gst-deploying-ai-solutions-for-data-mininggenerating-automated-notices-driving-govt-revenues/110370822. 50 Arpit Gupta, GST: Deploying AI solutions for data mining, generating automated notices & driving government revenues, ETGovernment, (May 24, 2024), https://government. economictimes. indiatimes. com/news/technology/gst-deploying-ai-solutions-for-datamining-generating-automated-notices-driving-govt-revenues/110370822.
2.1 Bias that may be created by Artificial Intelligence (AI):
When Artificial Intelligence (AI) is being deployed in the systems, it has the possibility of creating unequal outcomes towards a specific group of people and this is referred to as bias in AI. 51 There are two instances that might lead to bias in AI systems, the first is with respect to the design of the AI model itself and second is with respect to the training data that are being used to build these AI models. 52 The algorithms used by AI have the ability to detect “historical biases or systemic disparities embedded within the data they are trained on”, because of which the outcome also has the possibility of reflecting those biases or disparities, for instance, even minor biases can lead to huge “discriminatory outcomes”.
53 Therefore, at times, these biases might exist even in the data which are being used in the AI algorithms and if that data favors a specific thing over others, the outcome of it will reflect these imbalances. Similarly, algorithmic bias might happen when the algorithms unintentionally cause bias even though the data might be neutral thereby leading to discriminating results. On the other hand, it has also been stated that “human bias, being referred to as the cognitive bias can seep into AI systems through subjective decisions in data labelling, model development, and other stages of the AI lifecycle and these biases reflect the prejudices and cognitive biases of the individuals and teams involved in developing the AI technologies”. 54 The Impacts of Bias in AI can be categorized under three headings, mainly discrimination, unfair decision -making, trust and reputation.
55 Discrimination could be with respect to “ethnicity”, “gender”, “age”, or “financial background” as biases in the AI systems have the capability of “exacerbating societal disparities” and not included few groups in the decision-
51 E&ICT Academy, IIT Kanpur, Understanding Bias in Artificial Intelligence: Challenges, Impacts and
Mitigation Strategies, (Jul 28, 2023), https://eicta.iitk.ac.in/knowledge-hub/artificial-intelligence/understandingbias-in-artificial-intelligence-challenges-impacts-and-mitigation-strategies/.
52 SAP, What is AI Bias? Causes, Effects and Mitigation Strategies, (Oct. 30, 2024),
https://www.sap.com/resources/what-is-ai-bias.
53 SAP, What is AI Bias? Causes, Effects and Mitigation Strategies, (Oct. 30, 2024),
https://www.sap.com/resources/what-is-ai-bias.
54 SAP, What is AI Bias? Causes, Effects and Mitigation Strategies, (Oct. 30, 2024),
https://www.sap.com/resources/what-is-ai-bias.
55 E&ICT Academy, IIT Kanpur, Understanding Bias in Artificial Intelligence: Challenges, Impacts and
Mitigation Strategies, (Jul 28, 2023), https://eicta.iitk.ac.in/knowledge-hub/artificial-intelligence/understandingbias-in-artificial-intelligence-challenges-impacts-and-mitigation-strategies/.
making processes, which also undermines the principles of justice and equality for all the individuals. 56 Similarly, bias in AI can also result in unfair decision-making process which in turn might affect people’s trust in the fairness and reliability on technology. Bias in the AI algorithms which are used in the administration of GST might lead to several detrimental effects on the taxpayers as well as the economy as a whole. One of the immediate effects will be the infringement of the fundamental rights of the taxpayers provided under the Constitution of India. 57 As these biases might result in the AI-driven results be more favourable towards a particular segment of taxpayers only. Another implication of bias in AI systems might be with regard to the trust and reliability in the GST administration system.
In tax administration, public trust and reliability plays a very important role as most of the compliance mechanisms are on the basis of self-assessment and filing of returns on a timely basis. The frequent issuance of notices, even for minor discrepancies leading to penal actions are not only affecting the compliance rate amongst the taxpayers but also encouraging unnecessary litigation and instigating the taxpayers to resort to tax avoidance and tax evasion strategies. Furthermore, a notable issue with regard to the integration of AI, which is also often overlooked by majority of the people today, is the issue of confidentiality of the information shared by the taxpayers and data privacy. Once an AI software is being deployed by a taxpayer, it ends up giving access to all the financial data related to its business operations.
This leads to the question of confidentiality of the information and is very concerning as highly sensitive information related to the taxpayer’s businesses are being shared not just with the tax authorities, but also with a third-party system involved in managing the AI platform or software. Therefore, it can be said that AI in the GST framework has created both opportunities as well as challenges for the taxpayers. Even though India is one of the largest and fastest-growing digital economy in the world and the technological infrastructure of India is trying to increase
56 E&ICT Academy, IIT Kanpur, Understanding Bias in Artificial Intelligence: Challenges, Impacts and
Mitigation Strategies, (Jul 28, 2023), https://eicta.iitk.ac.in/knowledge-hub/artificial-intelligence/understandingbias-in-artificial-intelligence-challenges-impacts-and-mitigation-strategies/.
57 The Constitution of India, 1950.
transparency and ease in the compliance mechanism under the GST system, challenges still exist particularly with respect to the implementation of these laws in the practical world.
CHAPTER 3: SURVEY ANALYSIS
The present paper is based on primary as well as secondary data. To collect primary data, a non-doctrinal research method was used, and secondary data was collected through various articles and journals on the Internet. This chapter of the paper aims to understand and analyse the responses received through the survey carried out as part of the research. A well-structured questionnaire through Google Forms was shared via WhatsApp and email
with different sections of people to collect data and understand their perceptions of the Goods and Services Tax Network (GSTN) and the use of artificial intelligence (AI) in administering GST. The summary of the responses received is also attached as Annexure 1. For conducting the survey, a set of 10 closed-ended questions was framed, and the respondents for the survey mainly comprise the law students who are somewhat familiar with the taxation system, practicing advocates mostly dealing with civil matters, and businessmen involved in small and medium-sized businesses.
3.1 Results of the Survey Conducted
The following pie charts aim to show a visual representation of the responses received: Fig. 1
In Fig. 1, out of 80 respondents, 39 respondents chose the option “somewhat familiar”, whereas
41 respondents chose the option “very familiar”.
Fig. 2 In Fig. 2, out of 80 respondents, 37 respondents chose “No”, whereas 43 respondents chose “Yes”. Fig: 3
In Fig:3, out of 80 respondents, 19 respondents chose “No”, whereas 61 respondents chose “Yes”. Fig: 4 In Fig:4, out of 80 respondents, 17 respondents chose “No” and 42 respondents chose “Yes”. However, 21 respondents chose the option “Not sure, because not familiar with the GSTN. Interpretation (Fig. 1 to 4): Fig. 1 indicates that the percentage of people who are “very much familiar” with the GST administration system is almost equal to the percentage of “somewhat familiar” people, and therefore, it can be understood that a basic level of familiarity with the administration structure exists among all the respondents. This could be due to the professional or academic exposure towards the GST laws or a general awareness amongst the people. From Fig. 2, it can be inferred that the majority of the respondents feel that the GST administration is transparent.
Furthermore, it can also be seen that a majority of the respondents are aware of GSTN (Fig. 3). This is important in order to understand their perception towards the GST administration system and analyse the responses received in the subsequent questions. Over 52. 5% of respondents feel that the GSTN has increased transparency in the tax administration and reduced the burden of compliance, whereas over 21. 3% of respondents feel that the GSTN has not enhanced transparency (Fig. 4). However, 26. 2% of respondents were
not able to answer this question owing to their limited knowledge and familiarity with the GSTN system. Fig: 5 In Fig: 5, out of 80 respondents, 19 respondents chose “No”, whereas, 61 respondents chose “Yes”. Fig: 6 In Fig:6, out of 80 respondents, 48 respondents chose “Tax Authorities” and only 5 respondents chose “Taxpayers”. However, there were 27 respondents who chose the third option i.e. “Equally Benefiting both”.
Interpretation (Fig. 5 & 6) Over 76. 3% of the respondents feel that the small businesses are at a disadvantage because of the tech-based compliance mechanisms. (Fig. 5). While at the same time, over 60% of respondents feel that the tax authorities are benefiting more from the digitalization of GST. Both of these figures are extremely important as it forms a major part of understanding the perception of people towards the GST administration system. It can be seen that the majority of the respondents are of the opinion that the lack of access to technology and difficulty in the understanding of the tech-based compliance mechanisms has made it difficult for the small businesses to comply with the GST laws. It is also important to note that only 6. 3 % of the respondents feel that the taxpayers are benefiting out of it, while the remaining 33.
8% of the respondents are of the view that the digitalization of GST is equally benefiting both the taxpayers and the tax authorities. Fig: 7 In Fig: 7, out of 80 respondents, 65 respondents chose “Somewhat familiar”, whereas only 15 respondents chose the option “very familiar”.
Fig: 8 In Fig. 8, out of 80 respondents, 25 respondents chose the option “No”, whereas 55 respondents chose the option “Yes”. Fig. 9 In Fig. 9, out of 80 respondents, a total of 53 respondents chose the option “No”, whereas only
27 respondents chose the option “Yes”.
Fig: 10 In Fig:10, out of 80 respondents, a majority of the respondents, i. e. , 70 respondents, chose the option “Yes”, however, only 10 respondents chose the option “No”. Interpretation (Fig. 7 to 10) It can be seen that the majority of the respondents, i. e. over 81. 3% of the respondents are only somewhat familiar with the integration of AI in the GST administration system, which implies they only have a basic understanding with regard to AI and its use for compliance mechanisms. While over 68. 8% of the respondents feel that the compliance burden has reduced for the taxpayers because of the real-time monitoring of the GST filings and transactions. However, it can be seen that 66. 3% of the respondents feel that the integration of AI in the GST administration and its use for compliance is not free from bias, and a majority of the respondents i. e. over 87.
5% respondents are of the opinion that human intervention is necessary in GST compliance to prevent any bias that may be created by AI. This is important in order to understand the awareness of the people with regard to integration of AI within the GST framework and whether it has made the administration system more transparent and accountable.
4.2 Analysis
My understanding and analysis of the responses received stems from a significant gap that exists between the intended goals and objectives of GST and the actual experience of the
taxpayers in complying with the GST laws. There is a strong perception amongst the respondents that even though digitalisation of GST and the integration of AI has been done with the intention of enhancing the overall transparency and efficiency of the GST administration system, its implementation has been quite challenging for the small enterprises. Furthermore, the understanding of most of the people today is that it is only the tax authorities who are very well equipped with the latest tools to monitor the transactions and return filings and has ensured that there is strict compliance with the GST laws and all the compliance mechanism are timely being timely fulfilled and the taxpayers, on the other hand, are facing difficulties because of their limited accesses to resources.
Although, this has helped in detecting frauds and minimising tax evasion, it has also led to instances wherein minor and unintended errors have attracted scrutiny and huge penalties, thereby overburdening the tax authorities as well the judiciary. Furthermore, the respondents have portrayed a very strong opinion with regard to the integration of AI within the GST framework as most of them feel that it has reduced the compliance burden on the taxpayers. However, when asked about their perception towards bias in the AI systems, majority of the respondents feel that the outcomes which are determined by the AI algorithms might impact their businesses if at all there is no human intervention or oversight.
The respondents feel that human intervention is necessary in order to create a situation wherein the data or information shared by the businesses are protected and limited to the tax authorities only. Therefore, the survey analysis helped in understanding the perception of people towards the GST administrative system and gave a practical insight and an in-depth understanding of the issues outlined in the previous chapters.
CONCLUSION
The introduction of Goods and Services Tax (GST) is one of the most important transformative developments that has shaped the Indirect Tax landscape of our country today. The primary objective of conducting research in this area of taxation law was to understand how the administrative system has been revolutionized by technological advancements such as the GSTN, e-invoicing, e-way bills, input tax credit, and the artificial intelligence.
The GST system today is already heavily relied on technology, and the integration of AI within the system has further sparked debates with regard to data privacy and confidentiality of the taxpayer’s data and information. The survey analysis that was conducted as a part of the research has also shown that the administrative system cannot be solely dependent on the techbased mechanisms without any human intervention. Human intervention or human scrutiny is equally necessary to ensure that the outcomes are balanced wherein the tax authorities can properly execute their powers and responsibilities without having any detrimental impact on the taxpayer’s rights and obligations.
Therefore, although technology has created improvements in the ease of doing business for the taxpayers, and reduced the administrative burden on the tax authorities, it has also brought significant challenges with regard to its implementation because of the limited access to technology and the existence of digital divide in the country. As the taxpayers are still struggling to keep themselves at par with the tech-based compliance requirements, it brings us to the conclusion that the advantages of the technological developments are not equally beneficial to all, as the taxpayers are still at a disadvantaged position as compared to the tax authorities. There is still a significant gap between the goals and objectives that were intended while introducing GST in India and the actual experience of the taxpayers in complying with the GST laws.
BIBLIOGRAPHY
• Legislations:
1. The Central Goods and Services Tax Act, 2017
2. The State Goods and Services Tax Act, 2017
3. The Integrated Goods and Services Tax Act, 2017
4. The Central Board of Revenue Act, 1963.
5. The Constitution of India, 1950
• Books:
1. SINGHLA AKHIL AND GAUR KUMAR PAVAN, GST MINI READY RECKONER,
(Taxmann’s Publication (P) Ltd., 5th Edition 2024).
2. DATEY V.S, GST READY RECKONER, (Taxmann’s Publication (P) Ltd., 22nd
Edition 2023).
3. CHRISTOPHER JATIN A, HOW TO DEAL WITH GST SHOW CAUSE NOTICES
WITH PLEADINGS, (Taxmann’s Publication (P) Ltd., 4th Edition 2024). • Online Journals:
1. Krishan Arora et al, GST and Technology: All you need to know, [2022] 138
taxmann.com 551 (Article), 30 May 2022.
2. CA Mayank Mohanka, Digital Transformation of Indian Tax Administration: CBDT’s
‘Faceless e Assessments’ & CBIC’s e-Invoicing, [2019] 112 taxmann.com 234 (Article), 17 December 2019.
3. CS Charu Vinayak, Artificial Intelligence and Tax Compliance, Chartered Secretary,
April 2025.
4. Annapoorna, Goods and Services Tax: What is GST in India? Indirect Tax Law
Explained, ClearTax, (May 4, 2025, 8:30 PM), https://cleartax.in/s/gst-law-goodsand-services-tax.
5. Cleartax, Simplification of Indirect Taxes under GST Regime, (Jan. 23, 2025),
https://cleartax.in/s/cascading-effect-under-current-indirect-tax-regime.
6. Dr. B Ramaswamy, GST: New Legal Initiatives and its Implications, [2024] 158
taxmann.com 709 (Article), (Feb. 1, 2024).
7. Pavan Kumar. R, GST Administration Through the Goods and Services Tax Network:
A Critical Analysis, [2021] 127 taxmann.com 759 (Article), (Jun. 21, 2021).
8. Karthik Sundaram, GSTN- The New Network, National Law School of India Review,
Vol. 28, No. 2 (2016), pp. 114-122.
9. ClearTax, What is GSTN (Goods and Services Tax Network)?,(Jan. 08, 2025),
https://cleartax.in/s/what-is-gstn.
10. Dr. Sanjiv Agarwal, GSTN Common Portal – Salient features (Part-I), Tax
Management India.com, (Oct. 27, 2017).
11. Abhinandan Banerjee, GSTN: The Technological Backbone of India’s GST System-A
Comprehensive Guide, (Feb. 23, 2025), https://authbridge.com/knowledgecentre/glossary/gst-council-meetings.
12. Pramit Pratim Ghosh, Input Tax Credit (ITC) under GST, (Aug. 11, 2023),
https://tallysolutions.com/gst/gst-input-taxcredit/?srsltid=AfmBOoryWa6GVbu_wex7wIll2TUxClEJda1XkEmdeULJQKttA7 RLx5_W.
13. Renu Suresh, GSTN Introduces Case-Insensitive E-Invoicing System Effective June
1, 2025, India Filings, (Apr 08, 2025), https://www.indiafilings.com/learn/casesensitive-gst-einvoicing-system/.
14. GSTN- Goods and Service Tax Network, GST System Architecture and Rollout
Strategy, https://www.gstn.org.in/assets/mainDashboard/Pdf/GST%20Technology%20Archit ecture.pdf.
15. Amit Sarkar, How Technology Solutions Can Help Businesses Create Their GST
Framework,
BW
Businessworld, (Aug. 22, 2017), https://www.businessworld.in/article/how-technology-solutions-can-helpbusinesses-create-their-gst-framework-124472.
16. CA Surjeet Singh Negi, Invoice Management System- A paradigm shift in GST
Compliances, [2024] 166 taxmann.com 423 (Article), (Sep. 18, 2024).
17. Karan Kakkar and Pranav Aggarwal, Invoice Management System: Unlocking
efficiency for regular GST compliances? , [2024] 167 taxmann.com 275 (Article), (Oct. 10, 2024).
18. McCarthy. et al, A Proposal for the Dartmouth Summer Research Project on Artificial
Intelligence, (1955) AI Magazine, 26(2), 12-14.
19. Chris Smith, The History of Artificial Intelligence, University of Washington,
December 2006.
20. McCarthy, J., Dartmouth Summer Research Project on Artificial Intelligence, 1956.
21. V Sriram, Artificial Intelligence and Taxation: Redefining Compliance Standards,
TaxGuru, (Jul. 05, 2024), https://taxguru.in/income-tax/artificial-intelligencetaxation-redefining-compliance-standards.html.
22. CS Suparn Sekhri, Artificial Intelligence and Tax Compliance: A Paradigm Shift,
Chartered Secretary, April 2025.
23. N. Kerinc, M. Serrat Romaní, AI biases and its consequences on taxation, Maastricht
University Press, (Oct. 14, 2022).
24. CA Mayank Mohanka, Digital Transformation of Indian Tax Administration: CBDT’s
‘Faceless e Assessments’ & CBIC’s e-Invoicing, [2019] 112 taxmann.com 234 (Article), (Dec. 17, 2019).
25. CS Charu Vinayak, Artificial Intelligence and Tax Compliance, Chartered Secretary,
April 2025.
26. Dr. Keerthy V.S. et al, Artificial Intelligence Law & Ethics, MPP House, First Edition
2024, pg- 260-261.
27. The Times of India, Maharashtra using AI tools to detect GST frauds : CM, (Dec. 21,
2024), https://timesofindia.indiatimes.com/city/nagpur/maharashtra-using-ai-todetect-gst-frauds-cm/articleshow/116514635.cms.
28. Anju Sharma, Dr. Kapil Kumar, Detecting GST Fraud Through Machine Learning
Techniques, Journal of Emerging Technologies and Innovative Research (JETIR),
2021 JETIR March 2021, Volume 8, Issue 3.
29. RMPS & Co. Simplifying GST Reconciliation and Return Filing with AI: A Guide for
Businesses, (May 3, 2024), https://rmpsco.com/simplifying-gst-reconciliation-andreturn-filing-with-ai-a-guide-for-businesses/.
30. Abhinit Singh, Becoming a Future-Ready Accountant: How Generative AI is
Integrating into
GST
Management in India, (Sep 15, 2023), https://www.telegraphindia.com/edugraph/career/becoming-a-future-readyaccountant-how-generative-ai-is-integrating-into-gst-management-inindia/cid/1966404.
31. Arpit Gupta, GST: Deploying AI solutions for data mining, generating automated
notices & driving government revenues, ETGovernment, (May 24, 2024), https://government.economictimes.indiatimes.com/news/technology/gst-deployingai-solutions-for-data-mining-generating-automated-notices-driving-govtrevenues/110370822
32. E&ICT Academy, IIT Kanpur, Understanding Bias in Artificial Intelligence:
Challenges, Impacts and Mitigation Strategies, (Jul 28, 2023), https://eicta.iitk.ac.in/knowledge-hub/artificial-intelligence/understanding-bias-inartificial-intelligence-challenges-impacts-and-mitigation-strategies/.
33. SAP, What is AI Bias? Causes, Effects and Mitigation Strategies, (Oct. 30, 2024),
https://www.sap.com/resources/what-is-ai-bias.
ANNEXURE 1: SUMMARY OF THE RESPONSES
Indian Journal of Law and Legal Research Volume VII Issue IV | ISSN: 2582-8878
4. Do you think the GSTN has increased transparency in the tax administration and reduced the
burden of compliance?
80 responses
@ Yes
@ No
@ Not sure, because not familiar with the
GSTN
5. Do you think the small businesses are at a disadvantage because of the tech-based compliance
mechanisms?
80 responses
@ Yes @ No
6. With the rapid increase in the collection of tax and increasing number of tax disputes, who do you
think is benefiting more from the digitization of GST?
80 responses
@ Taxpayers @ Tax Authorities @ Equally benefiting both
Indian Journal of Law and Legal Research Volume VII Issue IV | ISSN: 2582-8878
7. How familiar are you with the use of Al in GST compliance?
80 responses
@ Very Familiar @ Somewhat Familiar
8. As Al tools enable real-time monitoring of GST filings and transactions, do you think the burden
of compliance has reduced?
80 responses
@ Yes @ No 68.8%
9. Do you think Al in GST compliance is free from bias?
80 responses
@ Yes @ No
33,
Indian Journal of Law and Legal Research Volume VII Issue IV | ISSN: 2582-8878
10. Do you think human-intervention is necessary in GST compliance to prevent any bias that may
be created by Al?
80 responses
@ Yes @No