India’s Income Tax Department is increasingly using artificial intelligence (AI) and data analytics to identify possible tax evasion, detect inconsistencies in returns and improve compliance. However, the growing use of these tools in assessments and appeals has raised questions about taxpayer privacy, accountability and the role of human judgement.
A commentary published by Mint on October 28, 2026, argues that the Central Board of Direct Taxes (CBDT) should establish a clear, publicly accessible framework governing how tax officials use AI. The central concern is that while automated tools can help identify risks, they should not replace the independent judgement required for tax decisions.
How AI Is Changing Tax Administration
The Income Tax Department uses technology under initiatives including Project Insight and the Income Tax Transaction Analysis Centre to analyse large volumes of financial information and identify transactions that may require further examination.
The systems can cross-reference information from sources such as banks, property registrars, stock exchanges, the GST Network, the Securities and Exchange Board of India and the Ministry of Corporate Affairs. Data from tax statements and returns can also be analysed to identify differences between reported income and other available financial information.
Machine-learning tools can help identify higher-risk cases for scrutiny, while automated workflows support aspects of faceless assessments and taxpayer communications. These methods may allow officials to focus their attention on cases that warrant closer examination rather than relying exclusively on manual checks.
However, a discrepancy identified by an algorithm is not necessarily proof of tax evasion. Differences may have legitimate explanations, making verification and an opportunity for the taxpayer to respond important parts of the process.
Concerns Over AI-Generated Assessments and Appeals
The commentary raises a further concern: AI is reportedly being used not only to identify potential compliance risks but also to examine taxpayers’ replies and help draft assessment orders or appeal decisions.
This creates a risk if AI-generated material contains incorrect legal references, misinterprets a judicial ruling or presents an inaccurate conclusion. The article describes an example in which an order relied on a tribunal ruling in a way that allegedly misrepresented the decision.
Such errors can have serious consequences for taxpayers if they influence decisions involving tax liabilities, penalties or appeals. AI-generated text may appear authoritative while still containing factual or legal mistakes, so its use requires careful review by qualified officials.
Why the CBDT Needs a Transparent AI Policy
The commentary calls for the CBDT to establish clear rules addressing several issues:
Transparency: Taxpayers should understand when AI has been used in processes that affect them.
Privacy and confidentiality: Financial and personal information must be handled securely.
Human accountability: Tax officials should remain responsible for decisions made in assessments and appeals.
Accuracy checks: AI-generated legal references, calculations and conclusions should be independently verified.
Grievance redressal: Taxpayers should have a clear way to report unauthorised or improper use of AI that affects their cases.
These proposals aim to distinguish the use of AI as an administrative aid from the delegation of legal decision-making to automated systems. The article argues that the CBDT should make its policy public and establish safeguards that protect taxpayers’ rights.
International Approaches Offer Possible Lessons
The commentary also points to approaches adopted in jurisdictions such as the European Union, the United Kingdom, Australia and Singapore, where AI governance frameworks address issues including transparency, security and human oversight.
The Organisation for Economic Co-operation and Development (OECD) has similarly emphasised the importance of reliable AI systems, accountability and meaningful human supervision. These principles are relevant to tax administration because decisions can affect a person’s finances and legal rights.
Mint
For taxpayers and businesses, maintaining accurate records and reconciling financial information before filing returns remains important as automated checks become more sophisticated. Professional income tax advisory services can help taxpayers understand compliance requirements and prepare appropriate responses to tax-related queries.
Balancing Enforcement With Taxpayer Rights
AI can help tax authorities process information more efficiently and identify cases that merit investigation. But its effectiveness depends on the quality of the data, the reliability of the systems and the judgement of the officials who use their outputs.
A transparent policy would help clarify where AI can assist, who is accountable for final decisions and how taxpayers can challenge errors. The key issue is not simply whether the tax department should use AI, but how to ensure that its use remains accurate, explainable and fair.