Arushi Yadav, HNLU Raipur, 3rd year
Artificial Intelligence (AI) is rapidly transforming India’s economy’s key sectors, driving efficiencies and innovation, while also giving rise to new competition challenges. In October 2025, the Competition Commission of India (CCI) released its landmark “Market Study on Artificial Intelligence and Competition”, examining how AI is reshaping market dynamics and what anti-competitive risks may follow. Conducted by the Management Development Institute Society (MDIS), this six-chapter report surveyed stakeholders, industry data and global literature to assess India’s AI ecosystem. It highlighted emerging issues in the market, from data dominance to opaque algorithms and proposes certain measures. The CCI report thus serves as both a diagnostic analysis and a call to action: identifying key AI-related competition concerns and proposing measures to foster a fair, competitive AI landscape in India.
The article examines the CCI’s findings and recommendations and offers a forward-looking perspective on India’s competition policy in the age of AI, highlighting the major concerns identified by the CCI, and discusses how India’s competition law can respond. It also compares enforcement approaches, drawing on global trends of ex-post and ex-ante regulation to suggest how India might proactively guard competition in AI markets. Throughout, it focuses on the Indian legal context, acknowledging recent legislative reforms and parallel initiatives such as the Digital Competition Bill, 2024. This analysis is aimed at legal professionals and students who wish to understand the CCI’s AI report in depth, as well as the emerging policy debates around AI and antitrust in India.
Competition Concerns in the AI Era
The CCI’s market study identifies several core competition issues in AI-driven markets. Foremost among these is the risk of algorithmic collusion. AI-powered pricing algorithms can independently monitor rivals’ prices and adjust their own, potentially coordinating prices even without any explicit agreement. Indeed, the report notes that “signalling algorithms and self-learning algorithms” enable rapid price coordination across markets, potentially stabilising tacit collusion in industries previously less prone to it. This “secondary algorithmic collusion” mimics cartel-like outcomes without a human agreement – a mode of collusion that traditional antitrust laws may not easily detect.
A second concern is that dominant firms might use AI to engage in exclusionary or exploitative practices. The CCI flags multiple issues like algorithmic price discrimination and opaque “black-box” decision-making, which may systematically favour incumbents or exclude competitors. For example, powerful platforms may train AI on massive data sets that smaller rivals lack, using the results to favour their own services or to personalise prices. Furthermore, AI also enables predatory pricing on a large scale and speed, as algorithms can undercut competitors dynamically and target vulnerable rivals more precisely than humans. These tactics echo classic antitrust abuses, but AI’s automation and scale can intensify their impact.
Additionally, AI thrives on large datasets and advanced infrastructure, which are often concentrated in big tech firms. Startups and smaller enterprises face steep obstacles: they may lack access to proprietary high-quality data, powerful computing hardware, skilled personnel and funding to train competitive AI models. In particular, data availability is the “most significant entry barrier” since AI models require vast, high-quality data, much of which remains in the hands of a few large corporations. The CCI’s report observes that the AI sector “tends towards concentration due to high upfront costs,” giving incumbents an “insurmountable competitive edge”. In practice, firms with privileged access can easily engage in self-preferencing, thereby reinforcing dominance and limiting alternatives.
Finally, AI raises transparency and consumer welfare issues. Many AI systems operate as “black boxes”, so consumers and competitors lack visibility into how decisions are made. Opaque algorithms can hide discriminatory pricing or unfair ranking, leading to unequal treatment of users or uncompetitive outcomes. The CCI cautions that where personalised pricing or targeted marketing is used, vulnerable consumers might be charged higher prices or excluded without realising it.
India’s Competition Law Response
How does India’s competition framework address these AI-era issues? The good news is that the Competition Act, 2002, is technology-neutral and already has broad prohibitions that can apply to AI-driven conduct. Section 3 outlaws cartels and anti-competitive agreements, and Section 4 prohibits abuse of dominant position. In principle, any AI-driven collusion or self-preferencing intended to restrict competition falls under the same prohibitions. CCI has increasingly engaged in digital markets: it has examined Google, Microsoft, Amazon, Meta and others in recent years, and issued orders on cases involving data accumulation and exclusionary practices. Importantly, the Competition Amendment Act, 2023, explicitly strengthened India’s toolset for the digital age. It penalised companies whose pricing algorithms kept their prices in lockstep (they used software agents to match each other’s prices on an online marketplace). This shows that even without a written “meeting,” automated collusion can violate anti-cartel laws. The amendment also creates new merger thresholds for large digital deals, empowering CCI to review big-tech acquisitions that might otherwise escape scrutiny.
At the same time, Indian authorities recognize the limits of a purely ex-post approach. Proving algorithmic collusion can be extremely challenging, as it requires evidence of explicit “meeting of minds”. The CCI report itself notes that proving “agreement or concerted practice” is hard when pricing decisions are autonomous. By contrast, newer regimes like the EU are exploring ex-ante rules to govern AI and big tech platforms. The CCI’s study advocates capacity-building and proactive engagement rather than immediate rules. For example, it recommends that enterprises conduct self-audits of AI systems for competition compliance by documenting algorithmic decision logic, monitoring outputs for inadvertent collusion or bias, and ensuring transparency in pricing algorithms. For its part, the CCI plans to strengthen its own “Digital Markets Division” with AI and data experts, hold industry conferences and advocacy workshops, and even establish an AI think-tank to stay ahead of tech trends. These measures reflect a cautious, iterative approach: fostering responsible AI adoption while preparing to intervene if needed.
At the policy level, India has yet to enact an AI-specific law; instead, it is working on voluntary guidelines and sectoral rules. However, India is also on the verge of a Digital Competition Bill (2024) that could introduce ex-ante obligations for dominant tech firms. The draft bill, after its enactment, would designate “Systemically Significant Digital Enterprises” (SSDEs) and bar them from self-preferencing, unfair bundling, or exploiting non-public data of others. Such rules, akin to Europe’s DMA, could directly tackle the AI-related abuses the CCI identified. In parallel, India’s pending Data Protection laws and non-personal data policies could shape AI competition by governing data access and portability. The CCI rightly notes that reducing entry barriers may require government action: proposals include building shared AI compute infrastructure and promoting open data platforms.
In sum, Indian competition policy is evolving to meet AI-era challenges. The CCI’s report doesn’t call for radical overhaul, but it does put AI squarely on the agenda. It signals that CCI is ready to combine enforcement with advocacy and inter-agency coordination, learning from global best practices while tailoring solutions to India’s context. As one analysis observes, India “cannot rely solely on ex-post enforcement” given AI’s structural risks a balanced mix of oversight and innovation support will be needed.
A Proactive and Collaborative AI-Competition Framework
While the CCI report provides a comprehensive baseline, certain forward-looking angles warrant emphasis. First, algorithmic accountability is crucial. Beyond self-audits, regulators and firms could develop shared standards or technical tools for “algorithmic impact assessments” in competition terms. For example, just as AI safety assessments are discussed for safety and privacy, one could imagine comparable processes for competition compliance, perhaps developed jointly by CCI, industry groups, and academia. This kind of cooperative governance (akin to India’s draft AI governance guidelines) could help demystify AI systems and detect collusive or exclusionary patterns early.
Second, the global dimension matters. AI markets are interconnected, and many leading AI models and platforms are multinational. The CCI report rightly mentions engaging international competition forums. India could benefit from aligning with initiatives like the Digital Competition Expert Panel (UK), OECD recommendations, or even the proposed US algorithmic transparency guidelines. Sharing data on algorithmic practices (in a privacy-compliant way) or participating in multilateral studies could help India spot emerging competition issues. For instance, if Indian startups struggle against a global AI firm, knowledge of how other jurisdictions handle self-preferencing or data access disputes could inform Indian enforcement.
Third, synergy with data policy is key. AI’s backbone is data, and India’s new Digital India Act (non-personal data framework), if enacted, may empower a Non-Personal Data Authority. The CCI and any future data regulator should coordinate to ensure competition doesn’t suffer. For example, if a dominant AI company controls massive non-personal datasets, can competition law require data-sharing or interoperability? The Competition Act’s existing Section 33(c) allows CCI to call data or information, but a formal “data trustee” regime (as debated by the Non-Personal Data Committee) could complement competition tools. Indeed, one proposal (SSRC) suggested harmonizing the CCI and Data Authority mandates. This intersection, ensuring open, fair access to non-personal data for AI training – may be the next frontier.
Finally, India’s push for indigenous AI (e.g. IndiaAI mission) could alleviate some concerns. By fostering homegrown large language models or open datasets, the government can reduce reliance on a few foreign tech giants. This is already happening (open-source models like Llama and others power much of global AI innovation). If Indian regulators encourage open source AI research (through funding, prizes, open data releases), the competitive landscape may become less concentrated. In effect, supporting public AI infrastructure and open innovation is itself a pro-competition measure.
In summary, tackling AI-competition issues will require both vigilant enforcement and creative policy. Beyond the CCI’s immediate recommendations, India should continue to craft a multi-pronged approach: encourage algorithmic transparency and auditability, align AI strategy with competition goals, and engage globally on AI norms.
Conclusion
CCI’s first AI and Competition market study marks a significant milestone in India’s regulatory evolution. It acknowledges that AI is not just a business tool but a market force that can reshape power dynamics. The report rigorously outlines the risks from clandestine price-fixing algorithms to data-based monopolies and urges both firms and authorities to act. For legal professionals, the key takeaway is that India’s competition law now squarely covers AI-driven conduct, and companies must integrate antitrust compliance into their AI deployment. Looking ahead, India is likely to pursue a hybrid regulatory model: using the Competition Act to police abuses, while also adopting forward-looking rules for digital platforms. The draft Digital Competition Bill, the Competition Amendment Act, and emerging AI governance guidelines together form a web of regulations that can address many of the CCI’s concerns. The CCI’s proposed “AI self-audit” framework, workshops and inter-regulatory coordination are practical steps to build an ecosystem of responsible AI. But success will depend on continued vigilance and dialogue: as the report implies, unchecked AI could entrench dominance, whereas careful oversight can instead spur innovation on a level playing field.