AI in Tax Governance in India and its Challenges, Benefits

India is increasingly leveraging Artificial Intelligence (AI) in tax governance, particularly through initiatives of the Income Tax Department such as Project Insight. While AI has significantly improved tax compliance, efficiency, and revenue mobilisation, it has also raised concerns related to privacy, bias, and accountability, making it a critical area for governance reforms.

Need of AI in Tax Governance

India’s tax system has long faced structural challenges such as a low tax-to-GDP ratio and widespread tax evasion, which constrain effective revenue mobilisation.

• Between 2001 and 2022, India’s average tax-GDP ratio remained around 16.36%, among the lowest for emerging economies.
• India loses around 4.3% of tax revenues due to tax evasion annually.

Traditional tax administration methods, being largely manual and reactive, have struggled to detect sophisticated financial irregularities in an increasingly complex and digital economy. In this context, the adoption of Artificial Intelligence (AI) has become necessary to shift towards a data-driven, proactive, and efficient tax governance system.

Use of AI in Tax Governance 

AI enables the integration and analysis of vast financial data, improves detection of income mismatches, and promotes voluntary compliance through behavioural interventions.

Project Insight (PI)

• Project Insight is a flagship initiative of the Income Tax Department launched in 2017 and made fully operational in 2019 to integrate AI and data analytics into tax administration.
• It aims to encourage voluntary tax compliance, reduce high-risk cases of potential tax evasion, make tax enforcement fairer and equitable, and reduce prejudice in tax enforcement.
• It develops a comprehensive financial profile of taxpayers by analysing data from multiple sources such as banking systems, GST filings, property transactions, and other high-value economic activities.

Project Insight functions through three interconnected components: 

• INTRAC (Income Tax Transaction Analysis Centre): The Income Tax Transaction Analysis Centre (INTRAC) is the analytical engine that utilises AI and advanced data analytics to process financial data from multiple sources, including banking and financial institutions, property and securities transactions, credit card and GST payments, and high-value purchases, to generate a 360-degree taxpayer financial profile.

• CMCPC (Compliance Management Centralized Processing Centre): It uses the insights generated by INTRAC to detect discrepancies and flag potential cases of under-reporting or misreporting. It prioritises cases based on risk levels, allowing tax authorities to focus on high-value or complex evasion cases.

• NUDGE Strategy (Non-intrusive Usage of Data to Guide and Enable): It represents a shift toward behavioural governance. Taxpayers receive SMS or email alerts when discrepancies are detected between reported income and financial activity. They are encouraged to revise returns voluntarily or provide clarification, reducing the need for coercive enforcement.

‹ ›

Impact of AI in Tax Governance

After receiving nudges, many taxpayers utilised the ITD’s updated-return feature to make voluntary changes to their original tax returns.

• Since 2020-21, AI-driven nudges have led to over one crore revised tax returns, generating approximately ₹11,000 crore in additional revenue.
• Out of the 19,501 taxpayers contacted by the ITD as part of a targeted Foreign Income and Assets NUDGE campaign, 62% of them corrected the information originally reported in their tax returns.
• 30,161 tax filers declared overseas assets totalling ₹29,208 crore and foreign income of ₹1,089 crore from cryptocurrencies or virtual digital assets.
• The NUDGE campaign covering 6.25 lakh taxpayers resulted in corrections of false claims for income-tax deductions amounting to ₹963 crore for political donations, and the payment of additional taxes to the tune of ₹410 crore. 
• The average time taken to process a tax refund has decreased from 93 to 17 days.
• Many advanced countries, such as Australia, Italy, the United Kingdom and the United States have successfully implemented AI-enabled platforms modelled on the PI and have generated additional revenue.

Benefits of AI in Tax Governance 

Deploying AI in tax administration has several benefits. 

• First, it can assist tax agencies in accurately assessing taxpayers’ risk profiles and identifying tax evasion. 
• Second, it enables tax administrators to prioritise tax evasion cases based on the size and sophistication of evasion. 
• Third, AI can automate routine tax administration tasks, freeing tax administrators to focus on those that require greater human judgment. 
• Fourth, AI can enhance taxpayer services by assisting taxpayers with filing correct tax returns, answering queries through smart chatbots, and preventing tax scams.

AI in Tax Governance Challenges

As India increasingly adopts AI-driven systems in tax administration, several critical operational, ethical, and institutional challenges have emerged that require careful attention.

• Data Provenance and Quality Issues: AI systems depend on the quality of data, and inaccuracies or complex financial patterns (such as variable incomes or joint family structures) can lead to false positives, forcing genuine taxpayers to justify their transactions.
• Algorithmic Bias: AI models trained on historical tax data may replicate existing socio-economic or regional biases, resulting in disproportionate targeting of certain groups or areas.
• Lack of Explainability and Due Process: The absence of transparency in AI decision-making makes it difficult for taxpayers to understand why they were flagged, undermining accountability and the right to challenge decisions.
• Data Privacy and Security Concerns: The extensive use of sensitive financial and personal data increases the risk of data breaches, cyber threats, and potential misuse of information.
• Institutional and Regulatory Gaps: The lack of an AI ombudsperson, mandatory audits, and public reporting mechanisms creates accountability deficits in AI-driven tax governance.

Ensuring Ethical AI-Based Tax Governance

To ensure that AI-driven tax governance remains fair, transparent, and accountable, India must adopt a robust ethical and regulatory framework.

• Human-in-the-Loop Approach: Critical decisions, especially those involving penalties or enforcement, must include human review to ensure contextual judgment and prevent over-reliance on automated systems.
• Establishment of AI Ombudsperson: An independent grievance redressal mechanism should be created to allow taxpayers to challenge AI-based decisions and address systemic errors.
• Algorithmic Transparency and Audits: Regular third-party audits and public disclosure of AI system performance are essential to ensure fairness, accuracy, and absence of bias.
• Ensuring Explainability and Due Process: Taxpayers must have the right to know the reasons for being flagged and access to simple mechanisms to respond, correct errors, or appeal decisions.
• Strengthening Data Privacy Framework: Robust data protection measures, including data minimisation, access control, and cybersecurity safeguards, must be implemented to protect sensitive information.
• Context-Sensitive AI Design: AI systems should be trained to reflect India-specific socio-economic realities, such as informal income patterns and family-based financial arrangements, to reduce false positives