Introduction

Competition regulation in India is currently operating on two separate tracks. On one side, the Competition Commission of India (CCI) is deliberately cautious with artificial intelligence, relying on advocacy and self-regulation. On the other, it is enforcing the Competition Act, 2002 against digital platforms with real penalties, settlements, and evidence-heavy orders. Understanding this dual approach is essential for technology companies, platforms, and in-house counsel assessing compliance risk in India in 2026.

Why India Is Regulating AI And Digital Markets Separately

India is targeting a US$5 trillion economy by 2027, and artificial intelligence is treated as a growth driver rather than a risk to be contained first. The domestic AI market is projected to reach US$131.31 billion by 2032, growing at a compound annual rate of 42.2%, with adoption spread across finance, healthcare, manufacturing, and logistics. Global capital has followed this growth, with multi-billion-dollar commitments announced by Microsoft, Google, and Amazon into Indian AI infrastructure.

Given the early stage of AI adoption, the government’s posture has been to build trust and capability first and regulate later. This was reinforced at the International AI Summit hosted by India in early 2026, where representatives from over 100 countries and more than 60 regulatory bodies concluded that safety, reliability, and misuse-resistance should take priority over antitrust intervention at this stage.

MeitY And CCI: Two Regulators, Two Mandates

AI governance in India is currently split between two institutions with distinct roles.

The Ministry of Electronics and Information Technology (MeitY) leads on safe, ethical, and responsible AI adoption. MeitY constituted a drafting committee in July 2025 to build a governance framework, drawing on existing Indian laws, global regulatory developments, and public consultation. The resulting India AI Governance Guidelines, titled “Enabling Safe and Trusted AI Innovation,” were formally released by the Press Information Bureau on 15 February 2026, timed to India’s AI Impact Summit 2026.

The Guidelines are structured in four parts. Part 1 sets out seven guiding principles, referred to as “sutras”: Trust is the Foundation, People First, Innovation over Restraint, Fairness and Equity, Accountability, Understandable by Design, and Safety, Resilience and Sustainability. Part 2 covers key issues and recommendations across six pillars : infrastructure, capacity building, policy and regulation, risk mitigation, accountability, and institutions. Part 3 lays out a short-, medium-, and long-term action plan, and Part 4 provides practical guidance for both industry (developers and deployers of AI systems) and government regulators.

On regulatory architecture, the Guidelines recommend three new institutions: an AI Governance Group (AIGG) to coordinate policy across ministries, a Technology & Policy Expert Committee (TPEC) to advise the AIGG on technical and legal questions, and an AI Safety Institute to lead standards development, testing benchmarks, and international cooperation. On the legal framework, the Guidelines note that many AI-related risks can already be addressed under existing law specifically the Information Technology Act, 2000, the IT (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021 and their 2026 amendments (which address AI-generated and deepfake content), and the Digital Personal Data Protection Act, 2023, which is enforced through the Data Protection Board of India. The Guidelines nonetheless call for a gap-analysis of these laws to address AI-specific issues such as liability across the AI value chain, application of data protection principles to model training, and use of copyrighted material in AI training.

Consistent with the “Innovation over Restraint” sutra, the Guidelines direct regulators to avoid compliance-heavy tools such as mandatory licensing unless necessary, and to prefer voluntary codes, standards, and regulatory sandboxes wherever real and present harm has not been demonstrated.

The CCI’s mandate is narrower than MeitY’s: it only steps in where AI-driven conduct is anticompetitive under the Competition Act, 2002. As part of its advocacy function, the CCI published its Market Study on Artificial Intelligence in October 2025. The study flags algorithmic collusion, bundling, self-preferencing, discriminatory pricing, entry barriers, reduced transparency, and market concentration as risks across e-commerce, healthcare, financial services, and logistics.

The CCI’s Self-Regulation Toolkit For AI Deployers

Rather than issuing binding rules, the CCI’s AI Market Study sets out a governance-style toolkit for any enterprise building or deploying AI systems in India. The recommendations include:

• Maintaining internal documentation and explainability of algorithmic design and outcomes, alongside external disclosure to users about AI-driven interactions and factors influencing promoted results.
• Testing algorithms in sandbox environments before public rollout, with defined triggers for human review of automated decisions.
• Collecting only the data strictly necessary for the AI system’s stated purpose.
• Conducting regular audits of AI outputs to detect unintended bias, predatory pricing, or discriminatory treatment of consumer segments.
• Training data scientists and product managers on competition law principles to build “compliance by design” from the earliest development stages.
• Assigning clear internal accountability, with senior management oversight and an impact-based approval process for high-risk or market-affecting AI deployments.

The CCI has included a full self-audit checklist and implementation process in the report. If an enterprise can demonstrably meet these standards, the CCI has indicated it is unlikely to intervene, and there has been no enforcement action against an AI deployer to date.

Ex Post Versus Ex Ante: How The CCI Is Approaching AI Regulation

CCI Chairperson Ravneet Kaur has publicly described the regulator’s current posture toward AI as a wait-and-watch approach, while noting that future intervention has not been ruled out. In parallel, the CCI is monitoring acquisitions and partnerships across the AI value chain for concentration risk and is exploring cross-border cooperation frameworks given the inherently global nature of AI markets.

On the question of ex ante (pre-emptive) regulation, the CCI has stated that its AI Market Study found that AI-driven markets carry risks similar to those seen in digital markets. This has fed into a broader legislative push. Following the enforcement of the EU Digital Markets Act (DMA) in 2024, India’s Ministry of Corporate Affairs (MCA), the ministry that governs the CCI released a draft Digital Competition Bill (DCB) in March 2025, after which stakeholder consultations concluded in July 2025.

In August 2025, the MCA issued a follow-up report on the CCI’s evolving role in regulating new technology markets. The report recommends that the DCB should:

1. Be enacted in a phased, evidence-based manner, focused on pre-emptively regulating “gatekeepers” or “systemically significant digital enterprises,” modelled on the EU DMA’s enforcement mechanism.
2. Introduce rebuttal mechanisms and clear designation thresholds so that growing Indian firms are not inadvertently captured.
3. Base these thresholds on findings from a formal market study rather than arbitrary criteria.

The DCB currently covers digital services such as search engines, browsers, social media, operating systems, cloud services, advertising services, and online intermediation. The MCA report additionally suggests bringing virtual assistants (such as Siri and Alexa) within scope and flags agentic AI systems as a likely future priority for the CCI, while cautioning against blanket prohibitions in favour of context-specific assessment. Whether AI-specific concerns such as data monopolisation for model training, algorithmic tying, and cloud compute dominance are eventually folded into the DCB, or regulated separately, remains an open legislative question. The India-EU Free Trade Agreement is also expected to strengthen momentum toward an ex ante digital markets framework.

Digital Markets: Active And Evidence-Based CCI Enforcement Under The Competition Act, 2002

Unlike its cautious approach to AI, the CCI’s enforcement in digital markets has been active. Digital markets accounted for close to 21% of all prima facie and final orders passed by the CCI in 2025, and 99 cases were brought before the CCI in this space between 2014 and 2024, with 64 ending in dismissal.

The First CCI Settlement: Google’s Android TV Licensing Case

A landmark development was the CCI’s first-ever settlement under Section 48A of the Competition Act. 2002 (as amended by the Competition (Amendment) Act 2023), read with the CCI (Settlement) Regulations, 2024. This closed a five-year investigation into Google’s Android TV licensing agreements, where original equipment manufacturers were required to bundle Google apps with the Play Store. Google proposed offering a standalone Play Store licence without mandatory bundling and waived certain compatibility requirements. The CCI accepted this on payment of a settlement fee of INR 202.4 million (approximately US$2.1 million), setting a precedent for future settlement applications under the new regime.

Dismissals Backed By Evidence, Not Assumptions About Dominance

The CCI has repeatedly clarified that dominance alone does not attract liability under Section 4 of the Competition Act, 2002, actual market foreclosure must be shown. Several 2025 decisions illustrate this evidentiary standard:

• Microsoft (Windows and Defender): The CCI dismissed claims that bundling Microsoft Defender with Windows OS, and related developer agreements, amounted to abuse of dominance, holding that users could freely disable Defender and install competing antivirus software, and that the antivirus market remained competitive.
• Google Play Store Policy Enforcement (Liberty Infospace and Zucol): The CCI found no prima facie abuse where Google terminated developer accounts under its “relation ban” policy for security and IP violations, holding that account-level termination is a legitimate deterrent against repeat offenders and that Google’s appeals process, including human review, was reasonable.
• BookMyShow: Although found dominant in online movie-ticket intermediation, the CCI held that differential treatment of single-screen cinemas versus multiplexes, advance payment terms, and lock-in clauses were commercially justified and mutually negotiated, and therefore not abusive. The CCI also reiterated that it does not regulate the fairness of commission fees.
• Zomato: A complaint alleging inflated pricing and mandatory platform fees was dismissed, as fees were non-discriminatory and features like tipping had visible opt-out options.
• Rapido: The CCI held that allegations concerning pricing and unlicensed vehicle use fell within the Motor Vehicles Act 1988, a special legislation that ousted the CCI’s jurisdiction over that specific conduct.

Data Privacy Recognised As A Competition Law Parameter: The Meta-WhatsApp Case

In a significant jurisdictional development, the National Company Law Appellate Tribunal (NCLAT) which hears appeals from CCI orders held in the WhatsApp LLC v. Competition Commission Of India matter that the CCI has jurisdiction to examine data practices that harm competition. The NCLAT confirmed that data privacy is a valid non-price parameter of competition and that mandatory, non-consensual data-sharing terms, such as WhatsApp’s 2021 privacy policy update, can constitute abuse of dominant position under the Competition Act, 2002.

The NCLAT partially upheld the CCI’s order, including a penalty of INR 2.13 billion (approximately US$22.8 million) on WhatsApp and Meta, holding WhatsApp dominant in the OTT messaging apps market on smartphones and finding that its data-sharing terms created entry barriers for rivals in display advertising. However, the NCLAT set aside the CCI’s finding that Meta had “leveraged” its dominance from messaging into advertising, holding that leveraging requires the same legal entity to exploit dominance across markets, and WhatsApp and Meta remain distinct entities. Both companies’ appeals to the Supreme Court, including a stay application, were dismissed after they agreed to comply with the NCLAT’s directions by 16 March 2026, with compliance certificates now pending before the CCI. The matter remains before the Supreme Court.

NCLAT Curtails Forward-Looking Remedies: The GPBS Judgment

In the same Meta matter, the NCLAT overturned the CCI’s blanket five-year prohibition on cross-data sharing between Meta entities where users had consented, holding that consent-based sharing should not be pre-emptively banned. A similarly significant reversal occurred in the Google Play Billing System (GPBS) judgment, where the NCLAT upheld the CCI’s underlying finding that Google abused its dominance by mandating GPBS for app developers, but struck down the CCI’s forward-looking, “gatekeeper”-style remedies, including data policy mandates and fee transparency requirements. The NCLAT held that Indian competition law requires proof of actual (ex post) violations, not pre-emptive gatekeeper-style regulation absent a codified ex ante framework.

The NCLAT also applied strict proportionality to penalties, reducing Google’s fine from INR 9.36 billion (roughly US$100.19 million) to approximately INR 2.16 billion (roughly US$23 million), a nearly 75% cut by basing the penalty solely on Play Store-specific “relevant turnover” in India rather than Google’s overall revenue.

The CCI Is Not A Price Regulator

Across the BookMyShow, Zomato, and Google Android TV decisions, the CCI has consistently held that it does not assess whether a company’s fees or commission structures are “fair,” so long as they reflect commercial reality and consumers retain a genuine opt-out. In the Google Android TV settlement, the CCI similarly declined to scrutinise the fairness of the standalone Play Store licence fee.

Sector-Wise Enforcement Trends To Watch

E-commerce: A long-running case involving Amazon and Flipkart, initiated in 2019 by small and medium smartphone traders alleging preferential treatment of select sellers, deep discounting, and FDI rule circumvention, remains active. The Supreme Court set aside NCLAT’s earlier findings in Flipkart’s favour in February 2026 and remanded the matter for fresh consideration, while Amazon’s own challenge failed and the investigation report against it is now before the CCI, with related proceedings stayed by the Karnataka High Court.

Online advertising: The CCI has consolidated multiple complaints against Google’s ad-tech practices, including allegations of tying its publisher ad server and supply-side platform with Ad Exchange, and linking YouTube ad inventory access to its own SSP. These have been joined with the ongoing News Publishers investigation, while a related complaint on search advertising was dismissed as duplicative of earlier cases.

Food delivery and quick commerce: Zomato and Swiggy remain under investigation following a National Restaurant Association of India complaint alleging exclusivity and price parity violations, with the Director General reportedly recommending findings of violation after a 30-month probe. Separately, quick-commerce players Blinkit, Zepto, and Swiggy Instamart face a complaint from the All India Consumer Products Distributors Federation alleging predatory pricing against offline retailers.

Key Takeaways For Businesses Operating In India

The direction of Indian competition law is now reasonably clear on two fronts. For AI, the CCI is prioritising self-regulation, documentation, and internal compliance-by-design over enforcement, but companies should not treat this as a permanent exemption, the CCI has explicitly reserved the right to intervene as AI markets mature, and the proposed Digital Competition Bill could eventually extend to agentic AI systems and virtual assistants. For digital platforms, enforcement is active, fact-specific, and increasingly reliant on economic evidence of foreclosure rather than presumptions based on size. The NCLAT’s insistence on proportional penalties tied to relevant turnover, and its rejection of pre-emptive gatekeeper remedies absent statutory backing, signals that Indian competition law currently operates on an ex post model pending the Digital Competition Bill’s enactment.

Conclusion

India’s regulatory approach in 2026 reflects a deliberate asymmetry: AI is being nurtured through guidance from MeitY and advisory studies from the CCI, while digital markets face sustained scrutiny under the Competition Act 2002, backed by settlements, penalties, and structural findings. For general counsel, compliance teams, and platforms scaling in India, the practical priority is twofold  implement the CCI’s self-audit toolkit proactively for any AI deployment, and ensure that pricing, data-sharing, and platform-access practices in digital markets are backed by clear commercial justification and evidence, since the CCI’s evidentiary bar for intervention continues to rise even as its enforcement activity remains high.

The issues examined in Apple vs CCI: Constitutional Proportionality & Revenue also illustrate the Competition Commission of India’s evolving enforcement approach towards AI-driven business models, digital platforms, and emerging technology markets.