Digital Twin Enterprises And Simulated Competition Distortions

Introduction

Digital Twin Enterprises are businesses that create continuously updated virtual representations of their physical operations, assets, customers, supply chains, employees, or entire commercial environments. A digital twin may combine operational data, IoT sensors, artificial intelligence, predictive analytics, simulations, and automated decision-making.

In competition law, the important question is not merely whether digital twins improve efficiency. The concern arises when an enterprise uses its digital twin to simulate competitors, predict their conduct, model market reactions, or test strategic outcomes in ways that reduce genuine competitive uncertainty.

A digital twin can therefore transform competition from an actual process of independent decision-making into a simulated environment in which firms can anticipate, coordinate, or strategically suppress competitive responses.

The principal competition concerns include:

1. algorithmic or simulated collusion;
2. coordinated pricing;
3. exclusion of competitors;
4. discriminatory access to infrastructure;
5. strategic use of competitor data;
6. predictive foreclosure;
7. automated retaliation;
8. self-preferencing;
9. raising rivals' costs;
10. manipulation of market entry conditions;
11. excessive concentration of data and computational resources; and
12. creation of feedback loops that make market power self-reinforcing.

1. Meaning of a Digital Twin Enterprise

A digital twin enterprise can be understood as an enterprise whose significant economic decisions are supported by a virtual model of its business and surrounding market.

The twin may represent:

• production facilities;
• logistics networks;
• warehouses;
• customers;
• suppliers;
• competitors;
• prices;
• demand;
• inventory;
• infrastructure;
• energy consumption;
• transportation;
• financial risks;
• advertising markets; or
• entire ecosystems.

For example, an online marketplace could maintain a simulated twin of its market containing:

consumers + sellers + competing platforms + prices + inventories + advertising responses + switching behaviour.

The enterprise could then run thousands of simulated scenarios before changing its actual prices or commercial conditions.

The competition-law difficulty arises when the simulated environment becomes sufficiently accurate that the undertaking can anticipate rivals' competitive behaviour and strategically exploit that information.

2. Simulated Competition Distortion

Traditional competition assumes that competitors make independent decisions under uncertainty.

A simplified model is:

Firm A → observes market → independently chooses strategy

Firm B → observes market → independently chooses strategy

Digital twins can alter this structure:

Market data → Digital Twin → Simulation of Firm A + Firm B + Consumers → predicted equilibrium → automated commercial decision

The enterprise may therefore cease competing simply by reacting to the market.

Instead, it may attempt to engineer the market toward a predicted outcome.

This creates what can be called simulated competition distortion.

3. Digital Twins And Algorithmic Coordination

The most serious concern is the possibility that digital twins facilitate coordination without a conventional cartel agreement.

Suppose competing enterprises possess highly sophisticated simulations capable of predicting:

• competitor price changes;
• capacity reductions;
• promotional campaigns;
• inventory movements;
• customer switching;
• likely retaliation.

Each enterprise may independently use its twin to arrive at the same commercially favourable strategy.

The resulting market may display:

• parallel pricing;
• reduced discounts;
• stable margins;
• reduced output;
• predictable responses to deviations.

The absence of direct human communication does not automatically eliminate competition-law concerns.

The central question becomes whether the conduct represents genuine independent competition or whether technology has facilitated concerted or coordinated behaviour.

4. Information Advantages And Competitor Simulation

Digital twins can produce enormous informational advantages.

A dominant enterprise may possess data concerning:

• competitors' inventory;
• customer behaviour;
• delivery capacity;
• supplier dependence;
• advertising expenditure;
• transaction-level pricing;
• production bottlenecks;
• geographic demand;
• product launches.

If the dominant enterprise feeds these data into its digital twin, it may simulate how a rival will respond to a commercial decision.

This can produce an informational asymmetry:

Dominant firm → highly detailed market simulation

Smaller rival → incomplete market information

The result may be a distortion of the competitive process even where the dominant enterprise does not formally prohibit competitors from entering the market.

5. Predictive Foreclosure

Digital twins may enable predictive foreclosure.

A platform could simulate:

"What happens if competitor X enters this geographic market?"

The twin might calculate that:

• prices should temporarily fall;
• advertising should increase;
• search rankings should change;
• exclusive contracts should be offered;
• suppliers should receive incentives;
• customers should receive targeted discounts.

The enterprise could then automatically implement the strategy.

This raises questions under abuse-of-dominance rules concerning:

• exclusionary conduct;
• predatory pricing;
• loyalty-inducing mechanisms;
• discriminatory treatment;
• refusal of access;
• tying;
• bundling;
• exclusivity; and
• raising rivals' costs.

6. Digital Twins And Market Definition

Digital twins may also complicate relevant-market analysis.

Traditional market definition frequently examines substitution.

A digital twin can model:

• hypothetical price increases;
• customer switching;
• cross-platform substitution;
• geographic responses;
• product substitution;
• multi-homing;
• network effects.

Thus, authorities could potentially use simulation technology to improve market-definition analysis.

However, the same technology creates a danger.

If a dominant undertaking controls the relevant simulation infrastructure, it may influence the assumptions used to model competition.

Consequently:

Control over the model can become control over the perception of the market.

This is particularly important where competition authorities rely on simulations supplied by the investigated undertaking.

7. Digital Twin Feedback Loops

A particularly important concern is the feedback loop.

Consider:

1. Digital twin predicts that consumers prefer Product A.
2. Enterprise increases Product A's visibility.
3. Consumers purchase more Product A.
4. The resulting data are fed back into the twin.
5. The twin concludes that Product A is even more dominant.
6. The enterprise increases Product A's visibility further.

The simulation therefore becomes partially self-fulfilling.

This can create:

prediction → intervention → market change → new data → stronger prediction → stronger intervention

Competition may consequently be distorted not because the original prediction was accurate, but because the enterprise used its market power to make the prediction come true.

8. Digital Twins And Self-Preferencing

A vertically integrated platform could create digital twins of competing sellers.

It may simulate:

• which seller is likely to expand;
• which seller could become a competitive threat;
• which seller depends on platform traffic;
• which products are most substitutable;
• which sellers could migrate to another platform.

The platform could then modify:

• rankings;
• recommendation systems;
• advertising placement;
• commission structures;
• access conditions;
• search visibility.

If the platform simultaneously operates its own competing service, the technology could facilitate self-preferencing.

The competition issue is particularly serious where the platform possesses both:

market information + gatekeeper control + competing downstream business.

9. Digital Twins And Raising Rivals' Costs

A digital twin can identify the precise points at which a rival is vulnerable.

For example, simulation might reveal that a competitor depends heavily upon:

• one supplier;
• one logistics provider;
• one cloud infrastructure;
• one payment processor;
• one distribution channel.

The dominant firm may then increase the rival's costs through strategically targeted commercial decisions.

This can constitute a technologically sophisticated form of raising rivals' costs.

The important point is that the discriminatory conduct may be highly selective rather than universally applied.

10. Digital Twin Market Manipulation

Digital twins can also facilitate dynamic market manipulation.

A platform might simulate millions of combinations of:

• prices;
• discounts;
• advertising;
• availability;
• ranking;
• consumer segmentation.

It can identify the strategy producing the greatest reduction in competitive pressure.

This creates a shift from:

competition by price and quality

toward:

competition by computational control of market conditions.

The competition authority may therefore need to examine not merely the observable price but the algorithmic architecture producing the price.

11. Relevant Case Laws

The following cases provide important legal principles for analysing digital-twin enterprises even though most pre-date the modern digital-twin technology itself.

1. United States v. Apple Inc. (2013)

The US litigation concerning Apple's conduct in the e-books market is important for understanding coordinated pricing.

The case demonstrates that technologically sophisticated markets can still be subject to traditional antitrust principles concerning:

• concerted action;
• price coordination;
• intermediary platforms; and
• market restructuring.

Relevance to digital twins

A digital twin capable of modelling competitor pricing could potentially make coordinated outcomes easier to achieve.

The important legal principle is that technological intermediation does not immunize coordinated commercial conduct from antitrust scrutiny.

2. United States v. Airline Tariff Publishing Co. (1994)

This case concerned mechanisms through which airlines communicated and adjusted fares through an electronic fare-publication system.

It is particularly relevant because the technology allowed competitors to observe pricing information rapidly.

Relevance

Digital twins take this concept considerably further.

Instead of merely observing competitor prices, a digital twin can:

• predict competitor reactions;
• simulate future prices;
• identify optimal responses;
• calculate retaliation strategies.

The case therefore provides an important conceptual foundation for examining technology-enabled reduction of competitive uncertainty.

3. United States v. Topco Associates, Inc. (1972)

The Supreme Court addressed territorial and customer restrictions among competing businesses.

The case remains significant for the principle that competitors cannot simply restructure markets among themselves in ways that eliminate competitive rivalry.

Relevance

A digital twin could theoretically be used to determine:

• geographic territories;
• customer allocation;
• market-entry strategies;
• competitor responses.

If such simulation facilitates market allocation or coordinated exclusion, conventional antitrust principles remain applicable.

4. FTC v. Actavis, Inc. (2013)

The Supreme Court considered the competitive consequences of settlements involving pharmaceutical patent disputes.

The case is important because the Court emphasised examining the economic realities and likely competitive effects rather than relying solely on formal legal characterisation.

Relevance

Digital-twin arrangements may similarly require authorities to look beyond labels such as:

"simulation," "optimization," "AI decision-making," or "predictive analytics."

The actual question should be:

What competitive effect does the system produce?

5. United States v. Microsoft Corp. (2001)

The Microsoft litigation is foundational for analysing technological ecosystems, exclusionary conduct and leveraging of market power.

The case examined Microsoft's use of its operating-system position to restrict competitive threats.

Relevance

A digital-twin enterprise could similarly leverage control over a core infrastructure or ecosystem to disadvantage competing products.

The modern analogue might involve:

• cloud infrastructure;
• operating systems;
• AI models;
• data platforms;
• digital marketplaces; or
• industrial simulation systems.

The case illustrates why control of an essential technological layer can generate competitive concerns in adjacent markets.

12. EU Competition-Law Cases

6. Google Shopping (Google and Alphabet v Commission)

The Google Shopping litigation is highly relevant to digital platforms and self-preferencing.

The case concerns Google's treatment of its comparison-shopping service within its general-search ecosystem.

Relevance to digital twins

Imagine a dominant platform maintaining a digital twin of every competing merchant.

The platform could determine:

• which rival presents the greatest threat;
• which rival requires greater visibility;
• how ranking changes affect competitors;
• how consumers switch between sellers.

The platform could then optimise its own downstream service.

The competition issue would therefore involve the interaction between:

data + simulation + gatekeeping + vertical integration.

7. Google Android (Google and Alphabet v Commission)

The Android case concerns Google's contractual practices surrounding mobile-device ecosystems.

Relevance

Digital twins can strengthen ecosystem control because the platform can model the consequences of:

• default settings;
• distribution agreements;
• application restrictions;
• interoperability conditions;
• competing services.

A digital twin could therefore become a sophisticated tool for maintaining ecosystem dominance.

8. Intel v Commission

The Intel litigation is important for the analysis of exclusionary rebates and the need to examine competitive effects.

Relevance

A digital twin could allow a dominant enterprise to identify precisely:

• which customers are contestable;
• which rivals are vulnerable;
• which rebates would prevent switching;
• which geographic markets require defensive incentives.

Such precision could make exclusionary strategies considerably more targeted.

The legal analysis should therefore focus on actual or potential foreclosure effects, not merely the sophistication of the technology.

13. Economic Mechanisms Of Distortion

Digital twins can distort competition through several mechanisms.

14. Digital Twin And Article 101 TFEU

Under Article 101 TFEU, the central concern would be whether digital-twin systems facilitate:

• agreements;
• concerted practices;
• information exchange;
• price coordination;
• market allocation;
• output restriction; or
• other restrictions of competition.

The fact that the decision is generated by software does not itself prevent liability.

A critical distinction is:

Independent algorithmic behaviour

Each undertaking independently develops its own optimisation model.

versus

Coordinated algorithmic behaviour

Competitors intentionally structure systems so that their algorithms anticipate or accommodate each other's conduct in a manner that substitutes coordination for independent rivalry.

The second situation presents considerably greater competition-law risk.

15. Digital Twin And Article 102 TFEU

For dominant enterprises, Article 102 concerns become even broader.

Potential abuses include:

A. Predatory pricing

The twin identifies the precise temporary price reduction necessary to eliminate a competitor.

B. Exclusive dealing

The twin identifies customers for whom exclusivity would be most effective.

C. Self-preferencing

The system predicts when preferential treatment will cause consumers to switch toward the dominant firm's product.

D. Refusal of access

The enterprise simulates the consequences of denying infrastructure or data access.

E. Discriminatory treatment

Different customers receive different commercial conditions according to predicted competitive threat.

F. Leveraging

The enterprise uses power in one market to reinforce its position in another.

16. Digital Twins And Data Concentration

A digital twin becomes more powerful as its underlying data become richer.

This creates a potential data-computational feedback loop:

More users → more data → better digital twin → better predictions → stronger market position → more users → more data

This may produce an endogenous barrier to entry.

A new competitor does not merely need:

• capital;
• technology;
• employees;
• customers.

It may also need the historical data necessary to train an equally powerful simulation system.

Therefore, competition authorities may need to consider whether data accumulation creates a durable barrier to entry.

17. Digital Twins And Network Effects

The digital twin can also intensify network effects.

Suppose a platform becomes more accurate because millions of participants use it.

The platform then offers superior:

• predictions;
• recommendations;
• logistics;
• pricing;
• fraud detection;
• demand forecasting.

Users consequently migrate toward the platform.

This increases data collection further.

The resulting cycle may be:

Scale → data → simulation accuracy → better service → more scale.

This can make digital markets particularly susceptible to tipping.

18. Digital Twins And Algorithmic Tacit Coordination

One of the most difficult issues is tacit coordination.

Suppose competing firms independently deploy digital twins.

Each twin learns:

"If I increase prices by 5%, my rival historically increases prices by 4%."

Over time, the models may converge upon stable strategies.

No explicit communication may occur.

The competition authority must then distinguish between:

1. lawful parallel behaviour;
2. conscious adaptation to observable market conditions;
3. algorithmic facilitation of coordinated behaviour; and
4. unlawful concerted practice.

This distinction is likely to become increasingly important as autonomous systems become more capable.

19. Evidentiary Problems

Digital-twin competition cases present unusual evidentiary questions.

Authorities may need access to:

• model architecture;
• training datasets;
• simulation parameters;
• reward functions;
• optimisation objectives;
• system logs;
• model outputs;
• automated decisions;
• human overrides;
• API records;
• version histories.

A major difficulty is that the relevant competitive strategy may never appear in a conventional document.

It may exist as:

code + model weights + simulation results + automated execution.

Consequently, traditional discovery and dawn-raid methods may need technological adaptation.

20. Liability For Autonomous Digital Twins

A central legal question is:

Who is responsible when the digital twin independently generates an anti-competitive strategy?

Possible candidates include:

• the enterprise;
• senior management;
• algorithm developers;
• data scientists;
• system integrators;
• third-party AI providers.

Competition law generally focuses on the conduct of the undertaking rather than treating software as an independent legal person.

Therefore, delegating a decision to an AI system should not automatically become a mechanism for avoiding competition-law responsibility.

The principle can be expressed as:

Automation may change how conduct occurs, but it does not necessarily change who bears responsibility for the undertaking's conduct.

21. Efficiency Defence

Digital twins can produce substantial legitimate efficiencies.

They can:

• reduce waste;
• optimise logistics;
• reduce energy consumption;
• improve production;
• predict equipment failure;
• reduce transportation costs;
• improve inventory management;
• increase product quality.

Therefore, competition law should not treat every simulation as suspicious.

The proper inquiry is whether the technology produces verifiable efficiencies that benefit consumers, or whether efficiency is being used as a justification for exclusionary conduct.

22. Competition Authority's Analytical Framework

A competition authority investigating a digital-twin enterprise could examine:

Step 1 — Identify the relevant market

Determine:

• product market;
• geographic market;
• platform/ecosystem boundaries.

Step 2 — Identify the digital twin

Determine:

• what the twin models;
• what data it receives;
• what decisions it controls.

Step 3 — Establish market power

Examine:

• market share;
• switching costs;
• network effects;
• data advantages;
• technological barriers;
• interoperability.

Step 4 — Examine simulation objectives

Ask:

• Is the system designed merely to optimise efficiency?
• Does it predict competitors?
• Does it identify exclusion strategies?
• Does it optimise for competitor elimination?

Step 5 — Examine outputs

Analyse whether the system produces:

• discriminatory prices;
• exclusion;
• coordination;
• foreclosure;
• self-preferencing;
• reduced output.

Step 6 — Examine implementation

Determine whether humans:

• approved;
• supervised;
• modified; or
• automatically implemented

the recommendations.

Step 7 — Assess competitive effects

Examine:

• entry;
• innovation;
• prices;
• quality;
• consumer choice;
• output;
• rival viability.

23. Regulatory Remedies

Potential remedies include:

Structural remedies

• divestiture;
• separation of business units;
• restrictions on vertical integration.

Behavioural remedies

• non-discrimination;
• access obligations;
• interoperability;
• data portability;
• restrictions on self-preferencing.

Algorithmic remedies

• independent audits;
• model documentation;
• logging requirements;
• simulation transparency;
• human oversight.

Data remedies

• data-access obligations;
• data portability;
• restrictions on combining datasets;
• interoperability requirements.

Governance remedies

• independent compliance officers;
• board-level responsibility;
• algorithmic competition assessments;
• periodic regulatory reporting.

24. Important Distinction: Simulation Is Not Automatically Anti-Competitive

It is important not to create a rule that:

digital twin = antitrust violation.

That would be economically and legally unsound.

Digital twins can intensify competition by allowing smaller firms to:

• predict demand;
• reduce costs;
• optimise production;
• improve supply chains;
• compete with larger incumbents.

The competition-law problem emerges where the simulation becomes an instrument for:

coordination, exclusion, foreclosure, discriminatory access, strategic retaliation, or durable market entrenchment.

25. Overall Legal Principle

The emerging principle can be stated as follows:

Competition law should examine not merely the observable conduct of digital-twin enterprises, but the computational architecture through which market outcomes are predicted, engineered and implemented.

The more autonomous and comprehensive the digital twin becomes, the more important it becomes to distinguish:

competition-enhancing simulation

from

competition-replacing simulation.

Conclusion

Digital Twin Enterprises represent a significant evolution in digital-market competition because they allow firms to construct virtual representations of markets and test commercial strategies before implementing them.

Their legitimate applications can generate substantial efficiencies. However, when combined with market power, large datasets, predictive algorithms, vertical integration and autonomous execution, digital twins may transform ordinary commercial strategy into a sophisticated mechanism for manipulating competitive conditions.

The principal competition-law risks are:

1. algorithmic coordination;
2. simulated price alignment;
3. predictive foreclosure;
4. self-preferencing;
5. raising rivals' costs;
6. strategic exclusion;
7. data-based entry barriers;
8. ecosystem reinforcement;
9. automated retaliation; and
10. concentration of computational control.

The cases involving Apple, Airline Tariff Publishing, Topco, Actavis, Microsoft, Google Shopping, Google Android and Intel demonstrate that courts and competition authorities already possess important principles concerning coordination, exclusion, technological leverage, vertical restrictions and competitive effects.

The novel issue presented by digital twins is that the competitive strategy itself can increasingly exist inside the simulation before it exists in the physical market.

Accordingly, future competition-law analysis will need to examine not only what an enterprise did, but also what its computational twin was designed to predict, optimise and cause.