From reactive battles to probabilistic, data-driven strategies, the next decade of tax litigation is being reshaped by artificial intelligence. If paired with human judgement and robust guardrails, this can potentially reduce disputes before they escalate.
On January 15, 2026, the day when the Supreme Court of India was scheduled to pronounce its verdict in the Tiger Global tax dispute, we experimented with an artificial intelligence (AI) model to assess its predictive capabilities on how the case might resolve. Drawing from patterns in prior rulings, judicial reasoning styles, and the trajectory of similar disputes through lower courts, the model assigned a 65% probability to one outcome based. When the judgment was finally delivered, the outcome aligned with the higher-probability outcome, though one success isn't necessarily a proven marker of accuracy.
The moment mattered not because AI was "right", but because it highlighted litigation's probabilistic nature, long recognised by experts, now more accessible via these tools. This effectively reframes litigation beyond the simple win-loss binary. Such tools could significantly influence tax strategies over the next decade, provided they integrate robust data and human oversight.
Reframing litigation and what it means for India Inc
Traditionally, litigation management in tax meant responding to notices, managing hearings, tracking appeals, and coordinating advisors. Technological upgrades in the form of digitised records, searchable databases, and electronic filings improved efficiency, but the approach remained largely reactive.
India's direct tax regime is now shifting from this legacy model through faceless assessments and appeals, the Vivad se Vishwas (2020) scheme, simplification measures, and, most recently, the new Income-tax law effective April 1, 2026. These reforms signal a structural reset, with the policy intent focused on reducing ambiguity, improving certainty, and lowering litigation over time, though Goods and Services tax (GST) challenges persist.
India Inc, today, faces surging GST notices amid reports of unprecedented volumes in FY25, largely triggered by automated reconciliations across returns, e-invoices, and third-party data.
The result is a disputes ecosystem where notices are issued faster than they can be meaningfully evaluated, and an ever-shrinking timeline to respond makes matters escalate before facts are fully understood.
AI adoption in tax becomes mainstream
Tax professionals are increasingly using AI to review notices, scan large volumes of case law, and draft initial submissions, all refined through professional judgment.
In parallel, tax authorities are deploying AI-driven analytics to scrutinise filings, reconcile data, and identify anomalies for closer examination. Assessments, across direct and indirect taxes, are becoming more data-led and pattern-driven.
This pits taxpayer AI against administrative AI, where deep data integration and human judegement provide a layer of accuracy and authenticity, though bias risk remains if unchecked
AI's real value lies not in drafting faster or summarising case law, but in using data to detect patterns, assess issue‑level predictability, and identify which disputes are likely to escalate versus resolve early. The power of AI-led analytics can group notices into archetypes and build institutional memory, thereby shifting litigation from a reactive backlog into a managed risk portfolio.
Complexity will take precedence in the next decade of tax litigation
With simpler issues likely to be resolved earlier under a more streamlined direct tax law, the disputes that survive will be the hardest ones, which are centred on treaty substance, permanent establishment, GST input tax credit, anti‑avoidance, and complex cross‑border attribution.
These disputes are rarely settled by black-letter law alone. The outcomes will be determined by how facts are interpreted, sequenced, and weighed, precisely the terrain where AI's multi-dimensional analytical capability adds significant value.
Capability without guardrails is a risk
As AI adoption accelerates, recent judicial observations underscore a critical truth: AI cannot replace human judgement in quasi-judicial processes.
Responsible use of AI in tax litigation requires a strong knowledge layer grounded in the subject matter, clear guardrails to prevent hallucination, mandatory human review, robust data governance, transparency of usage, and accountability. AI must remain an assistant, not an adjudicator.
What mature AI state may really look like?
A mature AI-enabled dispute management framework for taxpayers or tax authorities will eventually converge on three shared pillars.
In this direction, EY India has built the EY India AI Tax Hub, an integrated AI‑enabled framework grounded in deep tax functional expertise. It brings together a suite of specialised AI agents across tax research, compliance, and litigation, designed to work cohesively rather than in isolation.
Anchored in institutionalised data, this approach enables analytics and predictive insight to move to the core of dispute management, becoming the bedrock of how India Inc assesses risk, chooses battles, and manages tax litigation.
The question AI forces us to ask
The most consequential impact of AI will not be that it helps us draft faster or argue better. It will be that it forces a new question at the start of every dispute: should this dispute exist at all?
When taxpayers can see which positions are defensible and which are fragile, they will litigate more selectively. When tax administrations can distinguish meaningful anomalies from noise, they can focus on substance over volume.
AI can play a pivotal role in clearing judicial backlog not by "speeding up courts", but by reducing what reaches higher courts in the first place and by making what does reach them cleaner, narrower, and faster to decide.
The article is contributed by Manoj Rathi, Tax Partner, EY India and Gayatri Dutt, Senior Tax Professional, EY India.
Disclaimer: The opinions and views expressed in the article are those of Manoj Rathi, Tax Partner, EY India and Gayatri Dutt, Senior Tax Professional, EY India, and are provided only for general information purposes. The news and editorial staff of ET had no role in the creation of this article nor vouch for or endorse this content.
(You can now subscribe to our Economic Times WhatsApp channel)
