The Antitrust Implications of Big Data and Algorithmic Collusion: A Critical Analysis in the Enforcement of Competition Law in the Digital Economy

Paper Code: AIJACLAV28RP2025

Category: Research Paper

Date of Publication: Feb 10, 2026

Citation: Ms. Sneha K, “The Antitrust Implications of Big Data and Algorithmic Collusion: A Critical Analysis in the Enforcement of Competition Law in the Digital Economy”, 5, AIJACLA, 301, 301-319 (2025), <https://www.aequivic.in/post/the-antitrust-implications-of-big-data-and-algorithmic-collusion-a-critical-analysis-in-the-enforce>

Author Details: Ms. Sneha K, LLM (Corporate and Commercial Law), School of Law Christ (Deemed to be University), Bengaluru

Download PDF

Abstract

This paper critically analyses the antitrust implication of the digital economy including Big Data and algorithms collusion whereby it operates autonomously on a market behaviour through self-learning and complex algorithms. It seeks to determine and evaluate the material legal and technical gaps in the application of competition law especially where the conventional ideas of agreement and intent fail to cover the concept of the algorithmic tacit collusion. Available literature emphasizes how the current antitrust constructs which are based mostly on the analogue age find it difficult to cope with the form of obscurity independence and coordination that algorithms provide not to mention the presence of Big Data as a contemporary barrier to entry. The research examines the Competition Act 2002 and the Draft Digital Competition Bill 2024 with a doctrinal and analytic method and a comparison of the responses in the OECD European Union and United States. The results show that though suggested Indian reforms can handle hub-and-spoke cartels they cannot deal with autonomous algorithmic collusion because of implementation issues and the black-box problem. The paper ends with a proposal such as defining changes special purpose digital units and computer-based antitrust tools to protect healthy competition and consumer welfare in the digital economy.

Keywords:- Algorithmic Collusion Antitrust Competition Law Big Data and Digital Economy.

Paper Code: AIJACLAV28RP2025

1. Introduction

The digital economy of networking and intelligence has turned into an inescapable part of reality where big data analytics and advanced algorithms are no longer geared to act as tools functioning in markets but rather to create and design market mechanisms. Being defined as the new oil data has become one of the focal economic resources allowing companies to optimise their prices personalise their services and acquire a competitive advantage due to its volume speed and diversity.[1] Although these changes have improved efficiency and innovation they present significant threats to competition law systems that were designed in a prior and analogue economic environment.[2]

One of the main issues that have come about as a result of this transformation is the development of algorithmic collusion. High-level algorithms especially self-learning and machine-learning systems may monitor the market environment on their own modify tactics and reach collusive conclusions with no explicit interaction human will and even consciousness. This behavior would make the use of conventional antitrust enforcement extremely difficult since it depends much on the manifestation of concepts including but not limited to agreement intent and direct coordination. Even in the realms of data zero-price and multi-sided digital markets[3] the existing conceptualizations of market definition and dominance abuse have to be redefined. In addition the accumulation of large volumes of data by a few companies can serve as a strong entry barrier allowing data-intensive market leaders to cement market control.

The topicality of the study is supported by the fact that the current competition law regimes such as Indian Competition Act 2002 are still poorly equipped to combat algorithmic tacit collusion. In the different places where coordination arises due to the operation of autonomous units profit-maximising algorithms legislative and enforcement procedures fail to assign liability or intervene with effect. This study will thus assess the suitability of the contemporary legal regimes in solving these issues as opposed to alternative methods like the interpretive evolution institutional reform and computational antitrust methods.[4]

Considering that artificial intelligence and algorithmic systems are increasingly evolving there is a growing consensus among regulators and academics that the policy of competition readiness in the future needs to be created. This paper aims at determining constraints in current enforcement devices and the possibilities of adaptable legal frameworks legislative adaptation and aggressive detection instruments such as specialised digital units and algorithmic screening devices.[5] Finally the study also seeks to consolidate competition as a core value of the digital market economy as it will encourage efficient enforcement with the ability to inhibit undue concentration tacit collusion and irreversible damage to consumer welfare.[6]

1.1 Research Objectives

This research aims at critically examining the convergence between artificial intelligence (AI) algorithms Big Data and Competition Law with particular regard to algorithmic collusion in the digital economy. The project aims to unveil the legal and technological gaps in the current system of competition law by considering whether Competition Act 2002 and Competition (Amendment) Act 2023 are effective and whether the Competition Commission of India (CCI) can perform its institutional role in fulfilling the emerging antitrust needs.

In addition the study will conduct a comparative study on the international antitrust responses to algorithmic collusion embraced by the Organization for Economic Co-operation and Development (OECD) the European Union (EU) and the United States with the perspective of deriving appropriate best practices.

Lastly the paper provides a multi-pronged reform agenda of future-proofing competition law including the use of computational antitrust instruments ex-ante regulation and ethical AI norms to improve on effective competition law enforcement in digital markets.

1.2 Statement of Problem

The fast digitization of the global economy has essentially changed the competitive environment due to the prevalence of artificial intelligence (AI) and the application of analytics based on big data. Companies are becoming more and more dependent on self-learned algorithms that can handle large quantities of real-time data to automatically make adjustments to prices and tactics in reaction to market and consumer behaviour and frequently without human oversight. Although these technologies yield efficiency gains and benefits to the consumer they also present a huge challenge on the traditional antitrust enforcement. Traditional competition law systems have been formulated to deal with overt human motivated collusion based on the fact of communication agreement and deliberate coordination of firms. These assumptions are disturbed by algorithmic collusion especially in the tacit form. The behaviour of independent algorithms can mutually determine how to coordinate indirectly and can lead to anti-competitive results without any explicit agreement or human intent. This development has revealed a very crucial enforcement loophole with the current legal frameworks failing to identify assign or punish such behaviour.

The issue is particularly acute in India where competition law regime is not quite fit to align with the autonomy speed and opaqueness of algorithm-driven markets. Competition agencies at both national and international levels have recognized that conventional methods of detection and prosecution are not very effective when it comes to efficient coordination facilitated by AI. These issues are only worsened by the further increase of the market of digital platforms which poses a greater risk of distorted pricing less consumer choice and market dominance by firms with data-rich holdings. Unaddressed algorithmic tacit collusion endangers the consumer welfare market transparency and the reputation of the competition enforcement institutions. This may be followed by increased prices a low competitive force and lack of confidence in online trade. Even though current regulatory efforts point to the world transition into more adaptable competition rules the existing loopholes and contradictions make the systematic and targeted investigation a necessity. This paper thus considers how the three are connected namely algorithmic autonomy regulatory adequacy and enforcement effectiveness especially in regard to protection of fair competition and consumer welfare in the digital economy.

1.3 Research Methodology

The research methodology of this study is the doctrinal research in which the approach is mostly analytical. It entails a methodical analysis of the statutory provisions judicial cases and comparative regulatory measures on competition law in the digital economy. The main sources are the Competition Act 2002 Competition (Amendment) Act 2023 and the Draft Digital Competition Bill 2024 reviewed with the use of the competent international regulative tools including the Digital Markets Act and the progress of the relevant legislation in the European Union and the United Kingdom.

The reference material in the analysis includes secondary sources such as statutes regulatory frameworks judicial decisions policy documents and academic literature covering the subject of anti-competitive practices in digital and platform-based markets. These sources were identified using existing academic databases and governmental materials such as the top law and policy archives and websites of competition laws authorities like Competition Commission of India European Commission OECD and the Ministry of Corporate Affairs Government of India. The study is based upon qualitative legal analysis that is aimed at the clarity of the doctrines and comparative evaluation of the gaps and similarities and problems in the current competition law enforcement and to review the proposed regulatory responses based on both legal and policy approaches.

1.4  Research Questions

1. How has the evolution of big data and algorithmic pricing transformed traditional notions of collusion in digital markets?

2. What are the legal technological and statutory loopholes within the Competition (Amendment) Act 2023 in addressing algorithmic tacit collusion in India’s digital economy?

3. How does Indian competition law particularly the enforcement approach of the CCI fare in addressing algorithmic collusion compared to international antitrust responses of the Organization for Economic Co-operation and Development European Union and the United States and to extract the best practices adopted?

4. What regulatory or structural reforms are necessary to bridge the enforcement gap in detecting algorithmic collusion in the digital economy? Can computational tools and AI-assisted detection mechanisms play a meaningful role in real-time enforcement of competition law?

2. Literature Review:

There exists a large literature discussing how algorithmic pricing supports tacit collusion without having to explicitly coordinate human actions. The article Algorithms and Collusion: Bridging the Gap with Alternative Tools by Alexandra P. Mikroulea (2025)[7] shows that digital markets defined by network effects multi-sided platforms and data dependency violate the classical competition law assumptions especially in the definition of relevant markets and the reconciliation of competition law with data protection. Equally in Martin Spann et al. (2025)[8] in "Algorithmic Pricing: Implications of Marketing Strategy and Regulation" the profit-maximising algorithms may result in implicit collusion and price parity without an explicit agreement which makes detection and enforcement difficult. In support of this discussion in Algorithmic Pricing and Collusion: The Limits of Antitrust Enforcement Sumit Singh Bhadauria (2020)[9] asserts that autonomous pricing algorithms constitute a new form of collusion since no intent or communication is required creating huge enforcement gaps in the traditional antitrust framework. Jia Lee et al. (2018)[10] also support this issue by demonstrating through the interdisciplinary prism that the current legal regimes are structurally unprepared to approach autonomous tacit collusion.

A number of researchers concentrate on the use of big data as a market power and entry point in digital markets. In Big Data Analytics: A Cause of Concern about Competition? Priyadarsini T.P. (2019)[11] assesses the dangers of the anti-competitive behavior due to the concentration of data and recommends the improvement of the regulatory strategies. In Digital Economy Big Data and Competition Law Roberto Augusto Castellanos Pfeiffer (2019)[12] also advocates competition consumer protection and data protection policies to be coordinated and views privacy as a competitive parameter. Previously personal data and network effects complicated the determination of markets in two-sided digital markets and thus the definition of the market required a doctrinal rethink (Salvatore Maraventano 2016)[13]. Combined these articles highlight the importance of the data-driven dominance in the context of urging the conventional methods of evaluation of market power and competitive damage.

The platform-enabled coordination problem features strongly in the article by Abhishek Rudra (2023)[14] that studies the subject of Hubs and Spokes Arrangement through Algorithmic Collusion; Economic and Legal Implications analyzing the technology-mediated hub-and-spoke collusion in Indian competition law. In the study regulatory uncertainty complexity in enforcement and risks of over and under regulation are identified. In Antitrust Ramifications of Price Algorithms Yashmitha Manjunath (2021)[15] also criticizes the ambiguity that price algorithms introduce to the Indian law of antitrust and sees reduced market transparency increased barriers to entry and institutional constraints. In the article Digital Economy tryst with Competition Laws: Indian version Swetalana Rout (2020)[16] puts these issues into context within Indian enforcement practice where the traditional antitrust weaponry has proven insufficient to deal with non-price and platform based harms.

An emerging body of literature examines computational and remedial reactions to algorithmic collusion. In "Overcoming the Current Knowledge Gap of the Collusion and the Role of Computational Antitrust" Renato Nazzini et al. (2024)[17] stress the importance of the data-based screening of cartels factual research and long-term policy reform to counter the issue of algorithmic price coordination. In Remedies against Algorithmic Tacit Collusion Francisco Beneke et al. (2020)[18] write in favor of structural remedies based on the oligopoly theory claiming that a collusion-preventing intervention based on asymmetry can still be efficient. In the article Competition Law Policy & Regulation Digital Age Naratjaran et al. (2021)[19] propose a critical review of the role of antitrust responsibility in automated decision-making settings. Although not a legal science paper the article by Anurag Agrawal and D. Jaiswal (2012)[20] titled Machine Learning and AI When the Theory of Games Comes to Help offers valuable background theory of algorithmic strategy interaction that adds to the perspective of machine-based coordination.

Taken together these studies define that algorithmic pricing concentration in big data and platform-based coordination have introduced a change of fundamental mechanics of collusion and market power. Although scholars are generally in agreement about the insufficiency of traditional antitrust frames like agreement and intent there is less agreement on the enforcement models especially autonomous algorithmic collusion. This loophole clarifies why doctrinal development computational antitrust instruments and inter-disciplinary regulation methods are necessary- a problem that this study aims to resolve in the Indian and comparative antitrust environment.

3. Evolution of big data and algorithmic pricing in digital markets:

This aspect has significantly changed existing perceptions of coordination in digital markets as the evolution of big data and algorithmic pricing has redefined the mobility of the market increased the reach of the collective behaviour and provided challenges to classical antitrust ideas including agreement and intent.

Big data can be defined as massive multifaceted datasets that are produced in high velocity and generated by numerous sources and can be successfully managed and analyzed with the help of sophisticated algorithms and computational technologies. It is usually defined in volume velocity variety and value. Data and especially consumer data has become an important economic resource in the digital economy and is quite equivalent to intellectual property or oil in the earlier industrial stages.[21] At the same time artificial intelligence (AI) and machine learning have allowed algorithms to learn adapt and decide independently without necessarily being programmed or having a specific optimisation goal.[22]

The technological advances have produced tremendous efficiencies such as decreased cost of production and transaction high quality of products customised services and more effective allocation of resources. Although it is good to both companies and customers the commodity nature of the huge usage of big data and algorithm pricing has aroused competition legal issues especially in terms of collusion.[23]

Classically collusion referred to overt or explicit deals amongst companies to maintain prices above a competitive equilibrium and usually collusion occurs in oligopolistic markets which are characterised by few players entry barriers and transparency.[24] Although tacit collusion was conceptually acknowledged it was hard to prove under competition law regimes because of the evidence limitation. This traditional paradigm has changed drastically because of big data and algorithmic pricing.[25]

3.1. The Conventional Collusion Paradigms have been modernized to include:

i. Better Transparency on the Market and Interaction frequency:

The algorithms have a significant impact on the transparency of the market and the rate of interaction between the firms. In online markets the prices may be updated on a continuous basis and hence the strategies of competitors on pricing can be responded to immediately. Such greater transparency and quick responsiveness will help identify the non-parallel pricing patterns and make coordination more likely and sustainable even in markets that were traditionally not considered oligopolies.[26]

ii. Tacit Coordination and Expansion of the Oligopoly Problem Facilitation:

Algorithms allow companies to get coordinated results without the traditional contract and face-to-face communication and they make the oligopoly issue enduring to a wider group of market structures. Algorithms serve as the most efficient form of oligopolists making them more efficient at carrying out their oligopolist functions: they apply precision speed and rationality to market data and react to market changes which increases the competitive issues.[27]

3.2. New Algorithms of Algorithmic Collusion:

As noted by such scholars as Ezrachi and Stucke[28] there are several situations in which algorithmic collusion can take place:

i. Messenger Scenario:

Explicit cartel agreements that have been devised by humans are monitored and implemented by algorithms such as price tracking of competitors or coordinated price changes are carried out automatically. This situation is observed in the case of Amazon Marketplace where sellers used common pricing algorithms to match the prices.[29]

ii. Hub-and-Spoke Scenario:

Competing companies use a similar third-party algorithm or platform as a hub to indirectly direct pricing or market behaviour of the spokes. In the case of ETURAS an online bookings system communicated discount limits and the fact that the agencies had not separated themselves together with the awareness helped to find a concerted practice.[30]

iii. Predictable Agent:

Companies implement autonomous algorithms that act in predictable ways therefore minimizing strategic uncertainty and promoting tacit coordination due to simultaneous actions.[31]

iv. Self-Learning Algorithms (Autonomous Artificial Agents):

The most complicated case is the self-learning algorithm that is trained on reinforcement learning and deep learning methods. These systems automatically modify prices and plans by interacting in the market and can possibly come up with collusive results with no explicit human instruction. The black box approach of deep learning makes it difficult to identify where accountability can be attributed and poses major legal problems.[32]

3.3. Arguments against the Concepts of Agreement and Intent:

The use of algorithms in coordination and especially the use of autonomous agents directly disputes the classical antitrust conditions that there is a meeting of minds or an intentional human act. In a situation where algorithms learn to coordinate new and existing in the autonomous mode without human knowledge or communication the existing laws addressing the cartels operated by people fail to be effective at all.[33]

3.4. Pricing discrimination:

In areas of higher demand prices are set higher than in those with lower demand. Pricing discrimination can also be highly granular with the use of big data and algorithms. Companies are able to establish location online behaviour and personal characteristics and inferred willingness to pay prices. Even though that may lead to efficiency by maintaining price equality to the demand it presents concerns of diminished consumer surplus exclusionary and discriminatory consequences.[34]

The big data and algorithmic pricing are converged which has radically changed the dynamic of competitive forces by heightening and protracting the existing coordination mechanisms and re-conceptualizing the tacit and autonomous cooperation. These trends lead to the need to reconsider the current antitrust instruments to be more conscious of the aspects of the market including transparency and frequency of interaction to invent accountability mechanisms that would be appropriate in a more automated and algorithm-driven economy.

4. Legal technological and statutory loopholes within the Competition (Amendment) Act 2023 in addressing algorithmic tacit collusion in India’s digital economy

The growing dependence on the use of big data analytics and artificial intelligence (AI) in business operations has essentially transformed digital markets creating sophisticated antitrust issues that question the sufficiency of current competition law models.[35] One of the key issues is the existence of algorithmic tacit collusion when profit-incentive pricing algorithms create market coordination in the absence of human intervention or agreement. This would require critical analysis of the legal technological and regulatory loopholes in the Competition Act of India especially after Competition (Amendment) Act 2023 and enforcement powers of Competition Commission of India (CCI).[36]

4.1. Legal Loopholes in the face of Algorithmic Tacit Collusion:

In section 3 (1) and section 3(3) of the Competition Act 2002 horizontal agreements and arrangements are forbidden that lead to an appreciable adverse effect on competition (AAEC). Section 2(b) uses the term agreement as a broad term and admits of any and any understanding or concerted action formal or informal and legally enforceable or not.[37] Although this broad description can theoretically cover some varieties of algorithmic coordination there are still huge gaps in considering tacit and autonomous collusion.

i. Meeting of Minds and Intent:

The main challenge is to prove a meeting of minds or intent to act in concert that is that is at the core of the demonstration of an agreement or concerted practice.[38] Pricing in an instance of algorithmic tacit collusion can develop independently by observing market dynamics and behavior of its competitors without human design or even intention. Such algorithms are black box and as such it is challenging to attribute the intent of anti-competitiveness to human programmers or the algorithm itself.[39] When algorithms are not legal entities or businesses it becomes difficult to hold them liable as per the current definitions of legal entities.[40]

ii. Aggravation of the Oligopoly menace:

The algorithmic pricing improves the transparency of the market and allows quick and coordinated actions which means that the oligopoly problem can reach even so-called competitive markets. Collusive results can be reached by firms when they respond rationally and in parallel by algorithms without communication. The Competition Act fails to forbid parallel conduct caused by unilateral decision-making; therefore this has created a significant challenge in enforcement.[41]

iii. Minor Indian Jurisprudence:

In Indian jurisprudence there is limited jurisprudence on the algorithmic collusion. In Samir Agrawal v. ANI Technologies[42] (Uber case) the CCI refused to describe the pricing based on algorithms as a hub-and-spoke cartel referring to the lack of horizontal agreement between individual drivers. This ruling has been denounced due to its lack of recognition of the anti-competitive potential of algorithmic coordination that is facilitated by platforms.[43]

4.2. Competition ( Amendment ) Act 2023: Forward with Restrictions:

The Competition (Amendment) Act 2023 is an important law making advancement especially by broadening the meaning of Section 3 (3) to include enterprises or associations that actively promote horizontal agreements although these may not be involved in identical or similar trades.[44] This change enhances the capability of the CCI to deal with hub-and-spoke cartels particularly in markets of digital platforms.[45]

Nevertheless the amendment is still concentrated on the situations in which there is a personified agreement or facilitating hub. It fails to appropriately measure some less obvious types of algorithmic tacit collusion like predictable agent and digital eye scenarios in which algorithms both learn and optimise strategies independently and may lead to supra-competitive pricing without centralised coordination or human intervention. This is because the need to prove an agreement or meeting of minds still continues to be a significant limitation.[46]

4.3. Regulatory: 

There is a gap between regulatory frameworks and technologies. Outside of restrictions in doctrine there are serious technological and institutional constraints to India.

i.           Algorithmic Opacity: Self-learning and deep-learning algorithms are black-box systems that make they their decision-making processes largely unreadable to the developers and hence obstruct the ability to identify collusive strategies and intent.[47] [48]

ii.         Identification and Data Gathering: Algorithms collusion does not leave much communicative footprint as compared to traditional cartels. Preexisting investigative instruments and reliance on additional factors like parallel behavior might not be sufficient in the digital markets of the present time.[49]

iii.       Lack of Specified Expertise: As the enforcement will demand technical skills in AI data analytics and algorithmic systems which the CCI has limited institutional resources in.[50]

4.4. Competition Regime Proposals of Future-Ready:

In order to fill such gaps the multi-pronged futuristic policy strategy is needed (which will be further discussed in Section 6):

i. Shifting the Burden of Proof: Providing a rebuttable presumption of coordination in the case of increasing parallel behavior in digital markets which leads to AAEC.[51]

ii. Competition by Design: It is required that internalized corporate governance mechanisms and statutory compliance requirements be enforced to prevent the occurrence of collusive outcomes stemming out of competition.

iii. Transparency of algorithms: Enhancing regulated publication of algorithmic logic and source codes with the aid of leniency programmes and non-disclosure punishment.[52]

iv. Automated Competition Law: Integrating price-fixing and price-skimming per se prohibitions into algorithmic systems to stem out anti-competitive behaviour automatically.[53]

v. Data-Based Screening Tools: Making investment in sophisticated screening tools to preempt suspicious pricing trends.[54]

vi. Specialised Digital Wing: (Creating a digital wing of CCI) It entails the creation of a separate digital wing that has legal technical and data specialists.[55]

vii. Whole System and International Co-ordination: The integration of the cooperation with consumer protection agencies data protection agencies and international regulators.[56]

Even though Competition (Amendment) Act 2023 has strengthened the ability of India to combat facilitated collusion it is still not enough to combat independent algorithmic tacit collusion. To deal with this enforcement gap institutionally strengthened technologically knowledgeable and aggressive competition law measures are needed that transcend the classical antitrust paradigms.

5. The Enforcement of the Competition Law through Algorithms Collusion and Competition: India CCI as per the OECD EU and U.S. best practice in antitrust

The challenge of algorithmic collusion has become a major concern in the worldwide competition law forcing a jurisdiction like India Organization Economic Co-operation and Development (OECD) the European Union (EU) and the United States (US) to review their enforcement strategies.[57] A comparative study shows that there is a disparity in legislation readiness enforcement patterns and regulatory ideologies and it also identifies the best practices applicable to the regulation of algorithm-mediated coordination in digital markets.[58]

5.1 Indian Competition Law and Competition Strategy to be enforced:

India has not embraced an explicit framework that deals with the issue of algorithmic collusion specifically but the Competition Act 2002 is still being revisited considering the digital market trends. The Competition Commission of India (CCI) has also looked at algorithmic conduct in the industry of ride-hailing online travel and digital search engine. Nevertheless the enforcement remains a problem due to the software-based coordination especially the hub-and-spoke set up and the self-driving algorithmic pricing.

In Matrimony.com v. Google LLC[59] case CCI concluded that Google was unfairly and discriminately acting by manipulating search algorithms.[60] Similarly in Samir Agrawal v. ANI Technologies Pvt. Ltd.[61] (Uber) the CCI emphasized the impossibility of having a horizontal agreement between independent drivers under which the price was determined by an algorithm and concluded that acceptance of the price determined by an algorithm did not amount to a meeting of minds. This revealed a loophole in legislation in the section 3(3) of Competition Act which states that parties must be engaged in the same or similar trade.[62]

The Competition Bill 2024 proposed is aimed to alleviate this restriction by specifically expanding Section 3(3) to include hub-and-spoke conspiracies where an appreciable adverse impact on competition is assumed despite the facilitating hub being in another line of business. Such coordinated conduct is specifically recognised in India under this reform and a few other jurisdictions. Other actions proposed to increase the enforcement of CCI are compulsory corporate governance and antitrust compliance systems algorithmic disclosure systems automation of per se prohibitions use of data-driven screening systems and the creation of a specialised digital wing of the CCI.[63]

5.2. International Responses to Antitrust:

a. Organization for Economic Co-operation and Development (OECD)

The OECD has conducted massive research on collusion and algorithms and the algorithms have been found to raise the probability of tacit collusion in even traditionally hard-to-co-ordinate markets as a result of algorithmic pricing. It has warns that the current antitrust implements might not be adequate and suggests sector and market research modify merger control to consider the effects of algorithmic coordination and solutions to prevent the coding of collusive algorithms. OECD underlines the multidisciplinary enforcement policy by incorporating competition consumer protection data protection and technical authorities.[64]

b. European Union (EU)

Article 101 of the Treaty on the Functioning of the European Union (TFEU) is very clear that any agreement and concerted practices that limit competition are highly prohibited.[65] Another reason why the concept of concatenated practice applies to algorithmic collusion is the fact that the EU takes it very broadly paying attention to the collaboration in practice but not in the sense of an agreement.[66]

An example of this is the ruling of the Eturas case where the Court of Justice of the European Union found concertation since the market participants were aware of the limitations on algorithmic discounts and failed to dissociate themselves.[67] The decision established that online algorithms to coordinate or parallel behavior would be subject to the current antitrust regulations.[68] Even though EU competition law de facto assumes quasi-strict liability regime the officials do admit that it remains a continued challenge to assign responsibility in cases where self-learning algorithms are involved as an opaque black box.[69] The EU has reacted with sector inquiries expert reports merger control scrutiny transparency commitments behavioural and structural commitments and the commitment decisions to deal with concerns in proactive manner.[70]

c. United States (US)

The US antitrust enforcement has mainly centered on the situations when algorithms become a tool to enforce the express agreements commonly known as the “Messenger scenario. In United States versus The use of the pricing algorithms to coordinate prices by an online retailer David Topkins who acknowledged that the algorithms were used in this case illustrates how the Sherman Act is applicable in this case which is based on the Section 1 part of the Sherman Act.[71]

The US is a conservative and evidence based interventionist in fast changing digital markets focusing on innovation and case-oriented analysis.[72] There are still debates about the sufficiency of current laws in response to autonomous algorithmic collusion especially since it is problematic to establish intent and communication in this case. There are still issues around algorithmic opaqueness and algorithmic liability and the Department of Justice and Federal Trade Commission still investigate data-driven mergers and exclusionary behavior using a fact-intensive model of enforcement.

5.3 Best Practices Adopted

Comparative analysis shows that there are a number of best practices that can be transferred:

i. Adaptation of legislation and Liberal Interpretation: The proposed expansion of Section 3(3) in India is in line with the flexible interpretation of the concept of a concerted practice in the EU as it allows enforcing it against the idea of algorithmic coordination without explicit agreements.[73]

ii. Active Inspection of the Market: Continuous stripping of digital markets by the OECD and EU in the form of market research and sector investigations highlights the significance of the need to do so through continuous investigation in India.[74]

iii. Interdisciplinary Knowledge: The enforcement teams should include special teams of legal economic technical experts that will be necessary in the detection of complex algorithmic behaviour.[75]

iv. Mechanisms of Transparency and Accountability: EUNet influenced transparency and auditing measures as well as disclosure proposals by India are mechanisms to solve the issue of algorithmic opaqueness and enhance accountability.[76]

v. Effects-Based Analysis and Rebuttable Presumptions: The Eturas case by the EU in which the burden of proving intent is switched to a firm with knowledge of coordinated practices makes proving intent more difficult in an automated context.[77]

vi. Innovation vs. Regulation: The EU and OECD are more interventionist than the US which is based more on evidence-regulated restraint.[78]

The competition policy in India is also becoming more aware of the dangers of algorithmic coordination and suggesting both legislative and institutional solutions. A balanced future-oriented competition framework with its clear laws interdisciplinary enforcement active market surveillance and higher levels of algorithmic transparency can be achieved by India by selective adoption of EU and OECD best practices without undermining incentives to innovate.

6. Bridging the Enforcement Gap in Algorithmic Collusion: Regulatory Reforms and the Role of AI-Assisted Detection in Digital Competition Law

The high growth of the digital economy has made the application of big data analytics and artificial intelligence (AI) to be normalized which has led to the development of more sophisticated types of algorithmic collusion putting a burden on traditional competition law enforcement. These issues are based on fundamental algorithmic features such as transparency lack of provable intent or consensus among autonomous systems and the dynamism of digital markets. To resolve this enforcement gap a comprehensive reform agenda that includes regulatory and structural reforms and computational antitrust and ethical AI principles are necessary.[79]

6.1. Regulatory and Structural Reforms:

To curb algorithmic collusion it is compelling to review the current laws and institutional skills of regulatory agencies like Competition Commission of India (CCI).

a. Reconsideration of the Concept of Agreement and Liability:

The classic competition law is based on the demonstration of agreement or understanding of participants of the cartel. Pricing strategies can form as a result of self-learning systems even with no purpose or communication of the human actor in an algorithmic tacit collusion. Such algorithms are black-box and make it hard to attribute anti-competitive intent when it comes to either programmers or systems and undermines traditional evidentiary standards.[80]

b. Concertation and Reversal of Burden of Proof:

Based on the Eturas strategy formulated by the EU India can adopt the concept of rebuttable presumption of concertation where parallel behavior in digital markets produces a positive impact on competition that is appreciable (AAEC). This would put the weight on the enterprises to show that parallel results were as a result of unilateral business decisions and not a result of a coordinated process.[81]

c. Growing the liability of autonomous algorithms:

Although the Competition (Amendment) Act 2023 broadened Section 3(3) to include the hub-and-spoke cartels it still presupposes an identifiable agreement which is provided by a hub. More reforms are needed to cover the cases of autonomous artificial agents arriving at collusive equilibria on their own. An ethical strategy would place an affirmative responsibility on businesses to make sure that algorithms are compliant and makes the firms responsible of anti-competitive results produced by systems under their purview.

6.2. Ex-Ante Regulation and Markets Governance:

The use of ex-post enforcement alone is not sufficient in fast changing digital markets.

i. Competition by Design:

Competition by design entails firms integrating compliance with the competition law in the architecture of the algorithm and makes sure that the systems are structured against collusive effects and limit data interrelation which could lead to anti-competitive behaviour.

ii. Ethical AI Standards:

The risk of collusion can be prevented by incorporating ethical AI principles in the design of algorithms: fairness accountability and transparency. The politics of International policy even at the OECD highlights the need to strike a balance between competition and innovation objectives.[82]

iii. Merger Control Adjustments:

The role of algorithms in increasing the perceived coordinated effects has to be taken into consideration by merger control regimes. Trade that makes them more susceptible to algorithm-based collusion should be carefully inspected and structural solutions can also interfere with the coordination process by adding asymmetry and complexity.[83]

6.3. Institutional Reforms:

i. Specialized Digital Wing:

The CCI should create a special digital unit that includes engineers IT specialists data analysts and lawyers to oversee algorithm-driven markets hear complaints on the part of consumers proactively investigate and analyze relevant source codes (where needed).[84]

ii. Market Research and Investigations:

Frequent market research and industry-specific investigations are crucial to comprehend the workings of algorithms and what structural correlations ensure the successful practice of tacit collusion to shape enforcement priorities and legislation reform.

iii. Inter-Agency Coordination:

Since algorithms are multidimensional the efficient regulation means that the authorities in competition consumer protection data protection and technical institutions have to coordinate their work.[85]

6.4. Artificial Intelligence and Computerized Tools:

Real-time competition law enforcement can be greatly improved by using computational tools and AI-assisted mechanisms to deal with the data intensity and algorithmic opacities.

i. Detecting Algorithms Information-Based:

Economic and statistical analysis can be used to detect suspicious pricing patterns abnormal bidding behaviour and distorted market structures that are indicative of collusion using data-driven screening tools.[86]

ii. The concept of Algorithmic Transparency and Auditing:

Black-box concerns can be addressed by disclosure of algorithmic logic and pricing mechanisms as a matter of mandate especially when distortions in the market or complaints about the models are experienced in the market. The schemes of leniency-based self-reporting and non-disclosure penalties can make the scheme more compliant whereas algorithmic auditing such as sandbox testing can determine the vulnerability to per se violations despite technical complexity.[87]

iii. Automated Competition Law:

In cases where the infringement is clear per se like in price-fixing bid-rigging and market allocation competition law rules can be directly integrated into algorithmic systems so that they prevent anti-competitive behaviour automatically.[88]

iv. Artificial Intelligence Supported Investigations and Remedies:

AI tools may help the authorities to set competitive barometers experiment with market impacts and to assess solutions. The supervised simulation conditions may force firms to change algorithms to encourage competitive instead of cooperative strategies.[89]

Counteracting algorithmic collusion requires a technology-aware and futuristic competition legal framework. Such improvements to the enforcement gap as legislative refinement of agreement and liability institutional capacity active market regulation and the implementation of AI-assisted enforcement mechanisms in combination with competition by design transparency requirements and AI ethics can significantly reduce the enforcement gap without undermining healthy competition and innovation in digital marketplaces.

7. Findings & Suggestions:

The article concludes that big data and algorithmic systems have begun to radically transform the competitive environment in that they make possible new forms of horizontal coordination that surpass the conventional antitrust models. Algorithms enable collusion in a variety of ways with some being Messenger where the software serves to perform explicit human deals and others being more complicated Predictable Agent and Digital Eye scenarios where algorithms learn and use collusive strategies on their own. This independence greatly makes it difficult to implement central antitrust principles like agreement and intent since collusive results can be achieved without human intention and effort. As a result an enormous legislative void remains in the ability to deal with algorithmic tacit collusion inspired by profit-maximising algorithms that work without any communication. The oligopoly issue is also enhanced by algorithmic behavior as it enhances market transparency and coordinated responses are quick and easily generated even in markets that are considered to be competitive. The current antitrust regimes such as that of India are institutionally and technologically unprepared to effectively identify probe and convict such behaviour because of the black box nature of algorithms.

To combat these difficulties the paper highlights the need to have a multi-pronged reform agenda. To begin with the principle of agreement under the competition law will have to be re-tuned to include autonomous algorithmic coordination which may be rebuttably a presumption of concerted practice where parallel conduct caused by algorithm-driven actions is anti-competitive. Second institutional capacity has to be reinforced by specialised digital units as part of the enforcement agencies e.g. Competition Commission of India (CCI) with technical and legal expertise and analytical expertise. Third computational antitrust tools specifically data-based screening algorithms are needed to identify minor and hidden instances of collusion. Based on EU and OECD experience the paper also highlights the relevance of active market research merger regulation to concentrate on the impact of algorithms algorithmic transparency requirements and behavioural remedies requiring competitive programming that would protect market integrity and consumer welfare in the digital economy.

8. Conclusion:

This paper shows that the digital economy presents deep but subtle issues to competition law which had presumed the existence of rational human decision-making. The increased use of self-learning and machine-learning algorithms has facilitated tacit collusion and coordinated behaviour without human intent and action which makes traditional concept of agreement and intent less and less satisfactory. These issues are worsened by the accretion of a huge volume of data which generates high barriers to entry and clouds evaluation of the market power possessed by data-intensive incumbency.

The analysis indicates that the Indian competition law still has legal and technological weaknesses in addressing algorithmic tacit collusion and complicated hub and spokes structures. Even though the latest changes in legislation have exhibited positive signs in the direction of the facilitated coordination they are not enough to encompass the autonomous algorithmic behavior. Competitive analysis also brings out a variance in international strategies with the European Union using a more flexible and effects-based model United States keeping an apprehensive innovation focused approach and OECD promoting detailed market research and situational analysis.

In prospect the paper proposes a competition regime of the future based on computational antitrust technology algorithmic transparency specialised digital enforcement ethical standards of AI and constant market surveillance. Due to the lack of empirical research and adjudicated cases interdisciplinary research involving legal economic and computer sciences should be maintained over time to implement competition policies that ensure effective innovation consumer protection and competition in an ever more algorithmic market.

[1] N Koradia K Manokaran and Z Saeed ‘Algorithmic Collusion and Indian Competition Act: Suggestions to Tackle Inadequacies and Naivety’ in Steven Van Uytsel (ed) The Digital Economy and Competition Law in Asia (Springer 2021) 113 https://link.springer.com/chapter/10.1007/978-981-16-0324-2_6 accessed 1 September 2025.

[2] Kenji Lee ‘Algorithmic Collusion and Its Implications for Competition Law and Policy’ (2021) SSRN https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3213296 accessed 1 September 2025.

[3] Priyadarsini T P ‘Big Data Analytics: A Cause of Concern for Competition?’ (2019) NLIU Law Review 81 https://nliulawreview.nliu.ac.in/wp-content/uploads/2022/01/Volume-IX-Issue-I-81-107.pdf accessed 1 September 2025.

[4] Steven Van Uytsel Shubha Ghosh Mehra and Yasuhiro Uemura (eds) Algorithms Collusion and Competition Law (Edward Elgar Publishing 2023) https://www.elgaronline.com/edcollbook/book/9781802203042/9781802203042.xml accessed 1 September 2025.

[5] Case Comment Samir Agrawal v Competition Commission of India (2021) 8 RGNUL Financial and Mercantile Law Review 85 https://heinonline-org-christuniversity.knimbus.com/HOL/Page?handle=hein.journals/rlfladme8&div=10&start_page=85 accessed 1 September 2025.

[6] OECD Algorithms and Collusion: Competition Policy in the Digital Age (OECD 2017) https://www.oecd.org/en/publications/algorithms-and-collusion-competition-policy-in-the-digital-age_258dcb14-en.html accessed 1 September 2025.

[7] Alexandra P Mikroulea Algorithms and Collusion: Bridging the Gap with Alternative Tools (Springer 2025) https://link.springer.com/article/10.1007/s40319-025-01578-5 accessed 11 July 2025.

[8] Martin Spann and others ‘Algorithmic Pricing: Implications for Marketing Strategy and Regulation’ (2025) Journal of Retailing https://www.sciencedirect.com/science/article/pii/S0167811625000473 accessed 11 July 2025.

[9] Sumit Singh Bhadauria ‘Algorithmic Pricing and Collusion: The Limits of Antitrust Enforcement’ (2020) SCC Online http://www-scconline-com-christuniversity.knimbus.com/DocumentLink/bUp51lPd accessed 11 July 2025.

[10] Jia Lee and others ‘Algorithmic Collusion and Its Implications for the Law and Policy of Competition’ (2018) SSRN https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3213296 accessed 30 August 2025.

[11] Priyadarsini T P ‘Big Data Analytics: A Cause of Concern for Competition?’ (2019) NLIU Law Review 81 https://nliulawreview.nliu.ac.in/wp-content/uploads/2022/01/Volume-IX-Issue-I-81-107.pdf accessed 1 September 2025.

[12] Roberto Augusto Castellanos Pfeiffer ‘Digital Economy Big Data and Competition Law’ (2019) https://www.academia.edu/85661168/Digital_Economy_Big_Data_and_Competition_Law accessed 11 July 2025

[13] Salvatore Maraventano ‘Competition Law in Digital Markets in the Era of Big Data’ (2016) https://www.academia.edu/37179127/Competition_Law_in_digital_markets_in_the_era_of_the_Big_Data accessed 11 July 2025.

[14] Abhishek Rudra ‘Hubs and Spokes Arrangement through Algorithmic Collusion: Economic and Legal Implications’ (2023) Indian Journal of Law and Legal Writing https://heinonline-org-christuniversity.knimbus.com/HOL/Page?collection=journals&handle=hein.journals/injlolw9&id=1934 accessed 11 July 2025.

[15] Yashmitha Manjunath ‘Antitrust Ramifications of Price Algorithms’ (2021) SCC Online http://www-scconline-com-christuniversity.knimbus.com/DocumentLink/n6Bvr0GS accessed 11 July 2025.

[16] Swetalana Rout ‘Digital Economy’s Tryst with Competition Laws: The Indian Version’ (2020) SCC Online http://www-scconline-com-christuniversity.knimbus.com/DocumentLink/5Zg36eSJ accessed 11 July 2025.

[17] Renato Nazzini and others ‘Overcoming the Current Knowledge Gap of “Collusion” and the Role of Computational Antitrust’ (Stanford Law School 2024) https://law.stanford.edu/wp-content/uploads/2024/02/Algorithmic-Collusion.pdf accessed 9 September 2025.

[18] Francisco Beneke and others ‘Remedies for Algorithmic Tacit Collusion’ (2020) https://www.researchgate.net/publication/345335271_Remedies_for_algorithmic_tacit_collusion accessed 11 July 2025.

[19] N Natarajan and others ‘Competition Law Policy and Regulation in the Digital Age’ (2021) Indian Journal of International and Comparative Law https://heinonline-org-christuniversity.knimbus.com/HOL/Page?collection=journals&handle=hein.journals/iihcj14&id=313 accessed 11 July 2025.

[20] Anurag Agrawal and D Jaiswal ‘When Machine Learning Meets AI and the Theory of Games’ (2012) Stanford Education https://web.stanford.edu/~ajaiswal/AgrawalJaiswal-WhenMachineLearningMeetsAIandGameTheory.pdf accessed 11 July 2025.

[21] Lee and others (n 11).

[22] Priyadarsini (n 12).

[23] Mikroulea (n 8).

[24] Beneke and others (n 19).

[25] Lee and others (n 11).

[26] Natarajan and others (n 20).

[27] Rudra (n 15).

[28] Mikroulea (n 8).

[29] Nazzini and others (n 18).

[30] Natarajan and others (n 20).

[31] Mikroulea (n 8).

[32] Natarajan and others (n 20).

[33] Lee and others (n 11).

[34] Beneke and others (n 19).

[35] Priyadarsini (n 12).

[36] Koradia Manokaran and Saeed (n 12).

[37] Manjunath (n 16).

[38] Manjunath (n 16).

[39] Ana Pošćić and others ‘EU Competition Law in the Digital Era: Algorithmic Collusion as a Regulatory Challenge’ (2020) EU and Comparative Law Issues and Challenges Series (ECLIC) https://hrcak.srce.hr/ojs/index.php/eclic/article/view/11937/5913 accessed 3 September 2025.

[40] Priyadarsini (n 12).

[41] Koradia Manokaran and Saeed (n 12).

[42] (2021) 3 SCC 136

[43] Koradia Manokaran and Saeed (n 12).

[44] Koradia Manokaran and Saeed (n 12).

[45] Manjunath (n 16).

[46] Van Uytsel Mehra and Uemura (n 05).

[47] Poscic and others (n 40).

[48] Koradia Manokaran and Saeed (n 12).

[49] Koradia Manokaran and Saeed (n 12).

[50] Koradia Manokaran and Saeed (n 12).

[51] Koradia Manokaran and Saeed (n 12).

[52] Koradia Manokaran and Saeed (n 12).

[53] Koradia Manokaran and Saeed (n 12).

[54] Nazzini and others (n 18).

[55] Koradia Manokaran and Saeed (n 12).

[56] Van Uytsel Mehra and Uemura (n 05).

[57] Koradia Manokaran and Saeed (n 12).

[58]  Manjunath (n 16).

[59] Case Nos. 07 & 30 of 2012

[60] Soumya Hariharan and others ‘Antitrust Implications of Algorithmic Collusion’ (2020) NLS Business Law Review Blog https://forum.nls.ac.in/nlsblr-blog-post/antitrust-implications-of-algorithmic-collusion/ accessed 4 September 2025.

[61] (2021) 3 SCC 136

[62] Case Comment Samir Agrawal v Competition Commission of India (2021) 8 RGNUL Financial and Mercantile Law Review 85 https://heinonline-org-christuniversity.knimbus.com/HOL/Page?public=true&handle=hein.journals/rlfladme8&div=10&start_page=85&collection=journals accessed 4 September 2025.

[63] Koradia Manokaran and Saeed (n 12).

[64] OECD Algorithms and Collusion: Competition Policy in the Digital Age (OECD 2017) https://www.oecd.org/en/publications/algorithms-and-collusion-competition-policy-in-the-digital-age_258dcb14-en.html accessed 1 September 2025.

[65] Luca Calzolari ‘The Misleading Consequences of Comparing Algorithmic and Tacit Collusion: Tackling Algorithmic Concerted Practices under Article 101 TFEU’ (2021) European Papers https://www.europeanpapers.eu/e-journal/misleading-consequences-of-comparing-algorithmic-and-tacit-collusion accessed 9 September 2025.

[66] OECD Algorithms and Collusion (n 65).

[67] Poscic and others (n 40).

[68] Beneke and others (n 19).

[70] OECD Algorithms and Collusion (n 65).

[71] Sebastian Thomas ‘Autonomization and Antitrust: On the Construal of the Cartel Prohibition in the Light of Algorithmic Collusion’ in Sigrid Voeneky and others (eds) The Cambridge Handbook of Responsible Artificial Intelligence: Interdisciplinary Perspectives (Cambridge University Press 2022) 347 https://www.cambridge.org/core/books/cambridge-handbook-of-responsible-artificial-intelligence/autonomization-and-antitrust/6B026CF3CF0EC7092493D8E32AF3F782 accessed 9 September 2025.

[72] A P Grunes and M E Stucke ‘No Mistake About It: The Important Role of Antitrust in the Era of Big Data’ (University of Tennessee Legal Studies Research Paper No 269 2015) 3 https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2600051 accessed 9 September 2025.

[73] Ryan Calo Camerford and D Daniel Sokol ‘Antitrust and Regulating Big Data’ (2016) 23 George Mason Law Review 1129 https://scispace.com/pdf/antitrust-and-regulating-big-data-47snysps7d.pdf accessed 9 September 2025.

[74] OECD Algorithms and Collusion (n 65).

[75] Nazzini and others (n 18).

[76] Soumya Hariharan and others ‘Antitrust Implications of Algorithmic Collusion’ (2020) NLS Business Law Review Blog https://forum.nls.ac.in/nlsblr-blog-post/antitrust-implications-of-algorithmic-collusion/ accessed 4 September 2025.

[77] Hariharan and others (n 78).

[78] Nazzini and others (n 18).

[79] Beneke and others (n 19).

[80] Poscic and others (n 40).

[81] Nazzini and others (n 18).

[82] Calzolari (n 66).

[83] Nazzini and others (n 18).

[84] Jordyn Michaels ‘Pathways to the Light: Realistic Tactics to Address Dark Patterns’ (2022) 49(1) Rutgers Computer and Technology Law Journal 176 https://heinonline-org-christuniversity.knimbus.com/HOL/P?h=hein.journals/rutcomt49&i=186 accessed 10 September 2025.

[85] US Department of Justice Statement of the Antitrust Division on Its Decision to Close Its Investigation of the Internet Search and Paid Search Advertising Agreement between Microsoft Corporation and Yahoo! Inc (Press Release 18 February 2010) https://www.justice.gov/archives/opa/pr/statement-department-justice-antitrust-division-its-decision-close-its-investigation-internet accessed 10 September 2025.

[86]European Economic Association 990 1023 https://www.edegan.com/pdfs/Rochet%20Tirole%20(2003)%20-%20Platform%20Competition%20in%20Two%20Sided%20Markets.pdf accessed 10 September 2025.

[87] Urs Gasser ‘Recoding Privacy Law: Reflections on the Future Relationship among Law Technology and Privacy’ (2016) 130 Harvard Law Review Forum https://harvardlawreview.org/wp-content/uploads/2016/12/Vol.130_Gasser.pdf accessed 10 September 2025.

[88] Department of Consumer Affairs Guidelines for Prevention and Regulation of Dark Patterns (Government of India 2023) SCC Online http://www-scconline-com-christuniversity.knimbus.com/DocumentLink/4Vwr9bJf accessed 10 September 2025.

[89] Natarajan and others (n 20).