The dynamic development of the computer games market has led to a significant change in the monetization models used by game producers and publishers. The traditional sales model, based on a one-time purchase of a product by the consumer, has been largely replaced by solutions based on long-term user engagement and generating revenue through micropayments (microtransactions). Mechanisms known as loot boxes, consisting in the paid purchase of virtual packages with random content.

Although initially perceived as a means of enhancing gameplay, this mechanism is currently the subject of intense legal, economic, and social debate. It is increasingly being pointed out that the design of loot boxes utilizes psychological mechanisms similar to those that have been present in traditional gambling games for many years. The random nature of the reward, the uncertainty of the outcome, the “near miss” effect, the limited availability of certain items, and the use of dynamic animations intended to enhance the user’s emotional engagement all contribute to the blurring of the line between entertainment and gambling mechanisms.

Additional controversy stems from the fact that the vast majority of modern games using loot boxes are also aimed at minors. Unlike traditional gambling games, participation in these mechanisms does not require a specific age or meeting specific formal requirements. In practice, this means that random monetization mechanisms are also used by children and adolescents, who, due to their stage of psychological development, are particularly susceptible to the influence of behavioral design techniques and so-called dark patterns).

In recent years, the issue of loot boxes has ceased to be analyzed solely through the prism of gambling law. Regulations concerning consumer protection, digital services, and child safety in the online environment are gaining increasing importance. Discussions at the European Union level indicate that the future legal framework may be based not only on classic definitions of games of chance but also on instruments to combat manipulative practices and ensure a high level of protection for consumers using digital services.

At the national level, the problem remains equally relevant. Polish lawmakers have not yet decided to introduce a separate definition of loot boxes into the Gambling Act of 19 November 2009. This does not mean, however, that these mechanisms remain entirely outside the scope of existing regulations. In practice, administrative bodies assess each specific business model on a case-by-case basis, analyzing whether its design meets the statutory definition of gambling. At the same time, the development of the secondary market for trading in virtual items, particularly so-called skin gambling , is creating new interpretative challenges that the legislature did not anticipate when enacting the current regulations.

The changes introduced by the PEGI rating system in 2026 provided an additional impetus for reassessing the current regulations. The revised rules for classifying games with paid random mechanisms confirm the growing awareness of the risks associated with the use of loot boxes, especially for underage users. Although the PEGI rating is not a source of generally applicable law, its practical importance for the European market remains significant and may influence both the distribution of games and the future direction of legislative changes.

The purpose of this article is to analyze the current legal status of loot boxes under Polish and European Union law, taking into account recent regulatory changes, the practices of administrative bodies, and the experiences of selected European countries. Particular attention will be paid to whether the current regulations effectively protect consumers from mechanisms based on randomness and whether the current regulatory model meets the challenges of the modern digital economy.

The essence of loot boxes and their functioning models

The concept of a lootbox has not yet been defined in either Polish or European Union law. However, in the literature and by public institutions, it is generally accepted that a lootbox is a mechanism whereby the user obtains, for a fee or free of charge, a virtual package containing items whose contents remain unknown until opened. A characteristic element of this solution is randomness – the user has no influence on the item they receive, and the probability of obtaining individual rewards is determined by the game developer or platform operator.

At the definitional level, however, it should be emphasized that the term “lootbox” encompasses a wide variety of business models, the legal assessment of which cannot be uniform. Public debate often equates all mechanisms based on randomness with gambling, while from a legal perspective, individual solutions differ in both their economic structure and the degree of risk to the consumer. It is precisely this diversity that means that assessing the compliance of lootboxes with applicable regulations requires an analysis of the specific operational model, not just the presence of a random element.

The most classic model occurs in games where lootbox content is limited solely to cosmetic items, such as skins , animations, character outfits, or visual effects. These items do not affect gameplay or increase the player’s chances of success. They are intended solely for aesthetic purposes, allowing the user to personalize the appearance of their character or equipment. Such solutions were long considered relatively safe from a consumer protection perspective, but the development of secondary markets for trading virtual items has significantly changed their economic significance.

A good example is the Counter-Strike series, where weapon skins initially served only a visual purpose. Over time, however, a robust secondary market developed around these items, allowing them to be sold for real money. Consequently, the value of some virtual items began to reach several thousand, or even several dozen thousand, and in exceptional cases, several hundred thousand euros. In practice, this means that a randomly acquired item can have a tangible financial value, even though the game developer itself does not officially allow for its sale. The existence of an external market is one of the main arguments raised in the discussion on the classification of such mechanisms as potentially akin to gambling.

Loot boxes used in sports games, such as EA SPORTS FC Ultimate Team, are of a different nature. In this model, users purchase virtual packs containing player cards, coaches, or other team items. Unlike skins in games like Counter-Strike, acquired items directly impact gameplay. Acquiring rare players can increase a team’s competitiveness and improve player performance. Although the developer publishes information regarding the probability of receiving rewards in a given category, the user still doesn’t know the contents of a specific pack at the time of purchase, and the decision to purchase is based on a random mechanism.

Yet another model is found in so-called gacha games, extremely popular, especially in Asian markets and in the mobile gaming segment. This mechanism is based on randomization of characters, equipment, or other items necessary for further progression in the game. A characteristic feature of gacha systems is their close connection to long-term user monetization. Players are encouraged to repeatedly make micropayments to obtain exceptionally rare characters or items, the probability of obtaining which can be extremely low. The literature indicates that these solutions most fully utilize the mechanisms of behavioral economics and the psychology of addiction.

However, the most controversial are third-party platforms that enable the trading of virtual items and participation in games of chance that use items from video games as a form of currency. These services operate independently of game developers, leveraging the existing market for skins or other digital goods to organize mechanisms reminiscent of classic casino games. Users deposit funds or use their virtual items to participate in lotteries, roulette, duels, or other games based on chance. Unlike traditional loot boxes offered by game producers, the participant’s goal is not only to obtain a specific item but often to achieve a tangible economic benefit resulting from the possibility of reselling it.

From a legal perspective, the differences between the presented models are crucial. Not every mechanism employing an element of randomness automatically leads to its classification as gambling. Factors that should be assessed include, first and foremost, the potential for financial gain, the existence of a secondary market, the method of financing participation, the possibility of withdrawing funds, and the actual impact of randomness on achieving a specific outcome. In practice, this means a case-by-case analysis of the specific business model, rather than adopting a uniform classification for all types of loot boxes .

This approach is also reflected in the practice of many European countries. Both administrative bodies and courts are increasingly moving away from abstract assessments of the lootbox mechanism itself, focusing instead on analyzing their actual operation and impact on consumer interests. Consequently, the current legal debate no longer revolves around the question of whether lootboxes as a category should be considered gambling, but rather which monetization models justify their inclusion in a specific regulatory regime.

Loot boxes and the definition of gambling in Polish law

Assessing the compliance of lootbox mechanisms with Polish law requires, above all, an analysis of the provisions of the Gambling Act of 19 November 2009. Although the legislature has not yet decided to introduce a separate definition of lootboxes, this does not mean that these mechanisms remain outside the scope of applicable regulations. On the contrary, in practice, their legal classification depends on whether the specific operating model meets the criteria for one of the games specified in the Act.

The basic premise of the Gambling Act is to subject activities in which the outcome depends on chance to a specific regime, and the participant gains the opportunity to obtain a specific financial or material benefit. The Act does not use the term “lootbox” because it was enacted at a time when modern computer game monetization models were practically nonexistent. This necessitates a functional interpretation, taking into account the economic nature of the mechanism in question, not just its name or the technical solutions adopted by the game developer.

A key element of most loot boxes is undoubtedly randomness. The user making the purchase neither knows the contents of the package nor has the ability to influence the outcome of the drawing. However, the mere presence of a random element is not sufficient to classify a given mechanism as gambling. In practice, the nature of the prize received by the participant and the ability to assign it a real economic value are equally important.

This is where a fundamental difference between classic loot boxes offered by game developers and the mechanisms used by third-party platforms for trading virtual items becomes apparent. If the item obtained through a draw has a purely aesthetic function and cannot be legally exchanged for cash or used outside of the game environment, the arguments for classifying such a mechanism as gambling are significantly weaker. The situation is different when the item is de facto a property that can be freely traded on the secondary market, yielding a real financial benefit.

In practice, the greatest controversy surrounds so-called skin gambling. In this model, users use items obtained in-game as a means of participating in subsequent games of chance organized by third parties. Skins, which were originally purely cosmetic, are beginning to function as a kind of digital currency with measurable economic value. This mechanism leads to a situation in which participants risk losing items of real-world value in exchange for the opportunity to win an even more valuable reward. This structure bears a much greater resemblance to classic gambling games than the traditional micropayment systems used by game developers.

At the same time, caution should be exercised before drawing too far-reaching conclusions. The mere existence of a secondary market does not automatically mean that every loot box should be classified as gambling. From a legal perspective, a case-by-case analysis of the entire business model is necessary, including, among other things, the method of acquiring virtual items, the possibility of their resale, the role of the game producer, the scope of control over the trade in digital assets, and the actual economic significance of the rewards. Consequently, two mechanisms utilizing an identical element of randomness may be subject to entirely different legal assessments.

This position is also reflected in the practice of Polish administrative bodies. To date, there has been no established practice of automatically classifying all loot boxes as gambling. Authorities focus instead on analyzing specific business models and assessing whether they meet the requirements of applicable regulations. This approach reflects the nature of the Gambling Act, which uses functional definitions, leaving authorities considerable scope for assessing individual factual circumstances.

In this context, the practice of entering certain online platforms into the Register of Domains Used to Offer Gambling Games in Contravention of the Act has become particularly significant. However, such an entry does not mean that all platforms utilizing the element of randomness conduct illegal activities. Each decision is preceded by an assessment of the specific operational model of the given service. Consequently, it cannot be assumed that the lootbox mechanism itself has been deemed illegal in Poland. It is not the abstract technical structure that is being assessed, but rather its practical application.

Under current law, it seems more appropriate to ask not whether loot boxes per se constitute gambling, but which of their numerous operating models demonstrate characteristics that justify the application of the provisions of the Gambling Act. This approach avoids oversimplification and better reflects the reality of the digital market, where solutions with widely varying levels of risk to consumers coexist.

At the same time, it should be noted that even if a given mechanism does not meet the criteria for gambling within the meaning of the Act, this does not mean there is a lack of legal oversight. Modern regulations increasingly refer to consumer protection instruments, counteracting manipulative practices, and ensuring the safety of children using digital services. Therefore, analysis of loot boxes cannot be limited solely to gambling law. Regulations regarding consumer protection, digital services, and designing interfaces in accordance with fair trading principles are gaining increasing importance, and in many cases, they may prove to be a more effective tool for protecting users than traditional gambling law instruments.

Loot boxes as a challenge to consumer protection law and the regulation of digital services

Although the debate surrounding loot boxes has for many years focused primarily on gambling law, a shift in regulatory direction is now becoming increasingly apparent, both at the national and European Union levels. Contemporary challenges related to random mechanisms in computer games concern not only the classification of specific models as gambling, but also the compliance of the practices employed with the principles of consumer protection, the protection of minors, and the fair design of digital services.

This change is primarily due to the development of the digital economy. The mechanisms used by game producers are increasingly based not on traditional product sales, but on long-term user engagement and gradual increase in spending through appropriately designed psychological solutions. This phenomenon is referred to in the literature as behavioral monetization, or monetization that leverages knowledge from cognitive psychology and behavioral economics. The goal of such mechanisms is not simply to facilitate a purchase, but to create an environment that encourages users to make subsequent purchasing decisions impulsively or emotionally.

Of particular importance in this regard are so-called dark patterns, referred to in Polish literature as manipulative or deceptive design patterns. These design solutions exploit the workings of human perception and decision-making processes to induce behaviors that are beneficial to the entrepreneur, but not necessarily aligned with the consumer’s true interests. In the case of loot boxes, these can take a variety of forms – from counters counting down the time until the end of a promotion, through messages about the limited availability of specific rewards, to elaborate animations that enhance the emotional experience of opening the packages.

These mechanisms are not coincidental. Psychological research indicates that a reward system based on a variable ratio reinforcement schedule is one of the most effective ways to maintain long-term user engagement. This same mechanism has been used for many years in classic gambling games, where the unpredictability of rewards maintains a high level of motivation for subsequent attempts. In the case of loot boxes, this mechanism is transferred to the computer gaming environment and combined with an attractive audiovisual setting and the ability to immediately make another purchase.

From the perspective of consumer protection law, a crucial question is whether the use of such solutions could lead to a violation of traders’ obligations arising from provisions on fair market practices. It should be noted that contemporary EU regulations increasingly place greater emphasis not only on the content of information provided to consumers, but also on the design of digital interfaces. Therefore, the subject of assessment is increasingly not the product or service itself, but rather the architecture of the purchasing process and the impact of the interface on the user’s freedom of decision-making.

Underage users are particularly important here. Both the European Commission and the European Parliament have repeatedly stated that children using digital services require a higher level of protection than the average consumer. This stems from their limited ability to assess economic risk and their greater susceptibility to persuasive techniques used by businesses. In practice, this means that solutions acceptable to adult users may be deemed disproportionate or unfair if they are primarily targeted at children and adolescents.

The importance of this issue has increased following the entry into force of Regulation (EU) 2022/2065 on the Digital Single Market (Digital Services Act – DSA). Although this act does not explicitly regulate lootbox mechanisms, it establishes a number of obligations regarding the design of digital services and the protection of users from practices that may negatively impact their decision-making autonomy. In particular, the DSA emphasizes the need to ensure a high level of protection for minors and limit the use of solutions that exploit the vulnerabilities of specific user groups. This trend indicates that future assessments of the legality of lootboxes will increasingly be conducted not only through the lens of gambling law but also taking into account consumer protection standards applicable in the digital environment.

In parallel, the European Commission is working on a legislative package known as Digital Fairness, which aims to adapt EU consumer protection regulations to the realities of the digital economy. Issues under review include manipulative design patterns, interface design that exploits user vulnerability, and mechanisms that exert excessive psychological pressure during purchasing decisions. Although the legislative work has not yet been completed, the direction of the proposed changes clearly indicates that future regulations may also cover monetization mechanisms used in video games.

The European Parliament also highlighted the need to enhance the protection of minors in its resolution of 26 November 2025 on the protection of children online. The document indicated that mechanisms such as loot boxes, in-game currencies, and other systems based on chance should be subject to special scrutiny from the perspective of protecting children from addictive and manipulative digital practices. While the resolution is non-binding, it provides an important political signal indicating the direction of future legislative action at the European Union level.

A separate but crucial element of the modern user protection system is the PEGI age rating. Starting in 2026, this system will adopt a more stringent approach to games featuring paid random mechanisms, recognizing them as solutions requiring a higher age rating. While the PEGI rating is not a source of law and does not in itself determine the legality of specific monetization models, it reflects a growing consensus on the need to provide greater protection for minors from mechanisms that utilize randomness and behavioral design techniques.

The above circumstances lead to the conclusion that the future of loot box regulation will likely be shaped primarily by regulations concerning consumer protection and digital services, rather than solely by traditional gambling law instruments. While the Gambling Act focuses on the qualification of specific business models, contemporary EU regulations increasingly assess the design of digital services and their impact on the autonomy of user decisions. Consequently, assessing the legality of loot boxes in the future will require comprehensive consideration of both gambling law and regulations concerning consumer protection, digital services, and children’s rights.

Approach of selected European countries to regulating loot boxes – a comparative analysis

The lack of a uniform definition of loot boxes in European Union law has led individual member states to develop different models for regulating this phenomenon. These differences concern not only the legal classification of random-based mechanisms but, above all, the assessment of the risks loot boxes pose to consumers, especially minors. As a result, the European Union currently boasts both countries adopting a very restrictive approach and jurisdictions that prefer to analyze individual business models rather than create separate statutory regulations.

Belgium has taken the most stringent stance for many years. The Belgian Gaming Commission has determined that certain lootbox mechanisms meet the criteria for gambling if the participant pays a fee, the outcome depends on chance, and the reward represents a specific economic value. Consequently, some game producers have decided to remove paid lootboxes from the Belgian market or significantly limit their functionality. This solution was primarily preventative in nature and aimed at limiting children and adolescents’ exposure to mechanisms that utilize randomness as a monetization tool .

The Dutch experience was different. For many years, the Dutch supervisory authority took a similar stance to the Belgian one, deeming certain lootbox models to be in violation of gambling regulations. The dispute concerned one of the most popular monetization models used by Electronic Arts became the subject of years of administrative and court proceedings. However, the final rulings demonstrated that the classification of loot boxes cannot be based solely on the presence of an element of randomness, but requires consideration of the overall economic structure of the game, the method of trading virtual goods, and the actual potential for financial gain for the user. The Dutch experience thus highlighted the difficulties associated with applying traditional definitions of gambling law to new business models operating in the digital economy.

At the opposite extreme is the approach adopted in Poland. To date, Polish lawmakers have not decided to create separate regulations regarding loot boxes or introduce a statutory definition. This means that the assessment of individual models is based on applicable gambling regulations and an analysis of the specific factual circumstances. This approach provides administrative bodies with significant interpretative flexibility, but also limits predictability for businesses operating in the digital market.

The practice of Polish authorities indicates that a functional assessment of the specific business model is crucial. In the case of platforms enabling the use of virtual items as a means of participating in games of chance, authorities may apply the instruments provided for in the Gambling Act, including entry into the Register of Domains Used to Offer Gambling Games Contrary to the Act. However, this does not automatically mean that all loot boxes used in computer games are illegal. The Polish model is therefore based on an analysis of the economic impact of a given solution, not on an abstract assessment of the randomness mechanism itself.

An analysis of the solutions adopted in individual countries leads to the conclusion that what is becoming increasingly important is not simply classifying loot boxes as gambling, but rather protecting consumers from the psychological mechanisms that lead to excessive spending or compulsive behavior. Therefore, many countries are beginning to perceive the loot box problem as an issue that goes beyond traditional gambling law and requires the use of instruments appropriate to consumer law and digital market regulation.

This approach also aligns with actions undertaken at the European Union level. The European Commission and the European Parliament increasingly point out that the fragmentation of national regulatory models can lead to uneven levels of user protection in the digital single market. The global nature of game producers’ operations means that businesses operate simultaneously in multiple markets, adapting their business models to the most stringent requirements in force in individual countries. In practice, this means that future legal solutions will likely aim for greater harmonization of consumer protection standards at the EU level.

However, this doesn’t mean a complete ban on loot boxes is necessary. A much more likely approach would be to introduce requirements regarding the transparency of random mechanisms, the publication of actual reward probabilities, more effective age verification of users, and restrictions on the use of solutions that exploit the vulnerability of children and adolescents to persuasive techniques. Such a regulatory model would preserve the possibility of using micropayments as a legal method of financing computer games while simultaneously strengthening consumer protection.

From the perspective of Polish law, the experiences of other European countries have significant interpretative significance. They demonstrate that mechanisms operating at the intersection of gambling and digital services cannot be assessed solely through the lens of classic legal constructs developed for traditional casinos or lotteries. The development of the digital economy requires a more comprehensive approach, taking into account both the economic significance of virtual goods and the impact of interface design on consumer decisions. Consequently, the future model for regulating loot boxes will likely be based on a combination of instruments from gambling law, consumer protection law, and regulations governing digital services, rather than the exclusive application of one of these legal regimes.

Conclusions de lege lata and postulates de lege ferenda

The analysis leads to the conclusion that current Polish law does not allow for a uniform legal classification of all lootbox mechanisms. Despite the growing number of voices calling for the recognition of lootboxes as a form of gambling, the current legal status does not provide a basis for automatically subjecting this entire product category to the provisions of the Gambling Act of 19 November 2009. Each assessment requires consideration of the actual operation of the specific business model, the nature of the prize, the potential for further turnover, and the economic impact of user participation in the random mechanism.

This doesn’t mean, however, that the current regulations remain entirely insufficient. With respect to some models operating on the market – particularly platforms that use virtual items as a means of participating in games of chance or enabling their exchange for cash – current regulations may be applicable. The practice of administrative bodies to date demonstrates that the Gambling Act remains an instrument that helps counteract the riskiest forms of activity, especially when virtual goods begin to function as an equivalent of money or property.

At the same time, it’s important to note that the vast majority of modern loot boxes don’t pose a classic gambling law problem. Their primary purpose isn’t to organize games of chance in the traditional sense, but to create a monetization model that leverages psychological mechanisms that increase user propensity to make subsequent purchases. For this reason, the current regulatory debate is increasingly shifting from gambling law toward consumer protection law and the regulation of digital services.

It seems that this is precisely the direction that Polish lawmakers should also adopt. Attempting to classify all loot boxes as gambling would oversimplify the extremely diverse digital market. A much more rational solution seems to be creating separate regulatory obligations for mechanisms that utilize randomness, without the need for automatic application of the entire gambling law regime.

First and foremost, it seems reasonable to introduce full transparency into random mechanisms. Before making a purchase, users should be able to familiarize themselves with the actual probability of winning individual prizes, how the randomization algorithm works, and whether this probability remains constant for all participants. Such solutions already exist in some computer games, but currently they are primarily driven by voluntary decisions by businesses or requirements in specific foreign markets.

The second direction of change should be to strengthen the protection of underage users. In light of current psychological knowledge and the positions of EU institutions, there is little doubt that children are particularly susceptible to the influence of mechanisms based on a variable reward system. Therefore, it seems reasonable to consider limiting the ability of people under a certain age to purchase paid loot boxes or introducing mandatory parental control mechanisms to effectively manage minors’ expenses.

Regardless of the above, legislators should consider introducing more detailed regulations regarding third-party platforms enabling the trading of virtual items. It is this market segment that currently raises the greatest concerns from the perspective of consumer protection and compliance with the Gambling Act. In particular, situations in which items obtained in-game become a means of participation in subsequent games of chance or can be directly converted into cash require analysis. In such cases, the line between a digital service and gambling activity becomes significantly blurred, justifying the application of more restrictive oversight measures.

Obligations regarding marketing activities should also be a crucial element of future regulations. In practice, loot boxes are primarily promoted through influencers and online creators, whose audiences often include minors. While advertising collaborations in and of themselves cannot be deemed unacceptable, situations in which marketing messages exclusively emphasize the possibility of winning exceptionally valuable prizes, disregarding the actual probability of winning them, or employing techniques that could create unreasonable expectations among recipients regarding potential benefits, require special consideration. In this regard, both consumer protection regulations and regulations regarding the integrity of advertising messages may apply.

The issues presented demonstrate that the issue of loot boxes is not limited to gambling law. In fact, it exemplifies a much broader phenomenon involving the use of advanced digital design techniques to influence users’ economic decisions. Technological advancements increasingly render traditional private and public law frameworks inadequate for assessing new business models based on user behavior analysis and interface design that maximizes consumer engagement and spending.

Consequently, the future of loot box regulation will likely depend less on further expansion of the definition of gambling than on the development of European consumer protection standards in the digital environment. Regulations on the transparency of digital services, countering manipulative design patterns, and ensuring a high level of protection for children using the internet are becoming increasingly important . These instruments may become the primary tool for mitigating the risks associated with loot box operations in the coming years.

It should therefore be assumed that effective regulation of this phenomenon requires a multifaceted approach, combining instruments of gambling law, consumer protection law, and digital market regulation. Only such a comprehensive solution will achieve the right balance between the freedom of game producers to conduct business and the need to ensure a high level of protection for users, particularly children and adolescents, who remain most vulnerable to the negative effects of random-based mechanisms.

Artykuł Lootboxes in Computer Games – Between Gambling Law and Consumer Protection. Regulatory Analysis Against the Background of Polish and European Union Law. pochodzi z serwisu KIELTYKA GLADKOWSKI LEGAL | CROSS BORDER POLISH LAW FIRM RANKED IN THE LEGAL 500 EMEA SINCE 2019.

You see a video, a product catches your eye, and an “add to cart” button is already blinking in the corner of the screen. A few seconds later the order is placed, paid for, and on its way — all without leaving the app. That’s how TikTok Shop works: a closed-loop model in which the path from watching a piece of content to completing a purchase has been cut to the bare minimum. That very immediacy is its greatest strength and, at the same time, the source of its most serious concerns.

A platform that in 2017 drew around 72 million monthly users now attracts more than 1.5 billion — and between 11 and 15 million in Poland alone. TikTok has stopped being merely a place for entertainment and has turned into a powerful advertising and sales engine, combining influencer marketing, algorithmic personalization, and the emotional purchasing impulse. For businesses, it is a new and remarkably effective retail space. For consumers, it is an environment in which it grows ever harder to tell an authentic recommendation from an ad, or a fleeting enthusiasm from a considered decision.

That is precisely why TikTok Shop has landed at the center of lawmakers’ attention. While its model is not unlawful, the platform’s operation intersects with an entire web of regulation: from consumer law and the ban on “dark patterns,” through the EU’s DSA and DMA, data protection and safeguards for minors, all the way to advertising, media, and electronic communications law. This article shows where the convenience of one-tap shopping ends and the protection of the buyer’s free will begins.

What makes the TikTok Shop platform innovative?

Today, an effective marketing strategy is becoming an increasingly important priority for sellers. In Poland, monopolies in sales are relatively rare. Many competing companies operate in most industries. When purchasing products, consumers must choose from hundreds, or even thousands, of options offered by different brands. The scale of this phenomenon is evidenced by the fact that in the fourth quarter of 2025, over 2.9 million active businesses operated in Poland alone, the largest group of which were those associated with the retail sector. Meanwhile, buyers generally do not want to spend much time thoroughly analyzing goods available on the market. They are often guided by the opinions of other users, brand recognition, or media coverage. Therefore, in an information society based on the dynamic development of social media, tailoring advertising to current consumer needs and behaviors becomes crucial for running a business. Authentic and credible recommendations from trusted creators are becoming more important, and for many buyers, they are more persuasive than formulaic television commercials.

Considering the above arguments, many companies are making changes to their advertising strategies, for example, opting for influencer marketing. Online creators typically publish aesthetically and thematically consistent content that captures the interest of users with similar preferences and tastes. A business partnering with an influencer who shares similar values gains the opportunity to reach a large group of potential consumers, made up of the influencer’s followers. TikTok has become the dominant platform enabling the implementation of the marketing model described above. In 2017, the application had approximately 72 million monthly active users, and according to data from 2026, this number has increased to approximately 1.54 billion. In Europe alone, TikTok has already reached over 200 million users, and in Poland, the number ranged from 11 to 15 million. The average time spent on the platform is 70 minutes per day, which translates to approximately 35 hours per month. These statistics also indicate the continued growth of TikTok’s popularity, confirming the future of using social media for advertising and promotional purposes.

The development of influencer marketing significantly changed existing marketing practices, and its increasing prevalence led to the transformation of the TikTok app from a social media platform into an advertising system. The effective and profitable collaboration between media and advertising prompted the platform to take the next step in its development, combining these two sectors. Users were offered the opportunity to completely simplify the purchasing process. Previously, consumers only saw product advertisements, which attracted their attention and prompted them to search for sales offers. However, this pattern left them time to consider whether a purchase was truly necessary or necessary. It was also likely that, despite their interest in the product, they would eventually forget about the advertised product, and therefore their desire to purchase it.

The solution to the marketing strategy described above turned out to be a new feature presented by TikTok: TikTok Shop. The innovative nature of this tool is based on a closed-loop model, meaning the purchasing process takes place within a single app. Users first encounter content promoting a specific item. They then have the option to immediately purchase it by adding the advertised item to their shopping cart in the bottom corner of the app. TikTok acts as an intermediary for payment, shipping, and the entire order process. In this way, the app has evolved not only into a profitable advertising system but also an online store, becoming a marketplace platform that mediates payment, logistics, and order fulfillment.

The origins of TikTok Shop

Initially, the online shopping phenomenon developed through e-commerce. Its popularity contributed to the diversification of online sales into several business models: B2C, B2B, and C2C. The former involves a relationship between a business and an individual customer (examples include online stores such as Zalando, Zara, and IKEA). B2B refers to transactions between businesses, while C2C refers to sales between individuals, such as on platforms like Vinted, OLX, and Allegro.

These e-commerce models typically control the sales process independently. Their profits largely come from consumers who shop by searching for specific products they need. Entrepreneurs compete with each other through marketing activities aimed at convincing consumers of the quality of their products and building brand recognition.

In the next stage, the development of social media, and consequently influencer marketing, contributed to the emergence of a completely new type of buyer, one driven by impulse. Online creators present a specific lifestyle on their profiles in a significantly idealized form, which attracts the attention of their followers and becomes a role model. The desire to emulate the creator they follow can manifest itself both in their behavior and in the possessions they possess. The influencer thus becomes a person who inspires and encourages the purchase of a given product. Even if, from a rational perspective, the buyer doesn’t need the product, they often decide to purchase it under the influence of influencer marketing.

Additionally, a new branch of e-commerce has emerged, known as discovery commerce . This model relies on the discovery and purchase of new items while actively browsing social media. Highly advanced algorithms select content for users that aligns with their tastes or interests, in order to evoke certain emotions that then transform into a strong purchasing impulse. Social media platforms, recognizing this profitable sector, have contributed to the development of social commerce, including TikTok Shop. This solution capitalizes on users’ fleeting enthusiasm and allows them to complete their order without leaving the app. The entire process, from advertising content to payment and shipping, is handled by TikTok, which can limit the time available for rational purchase consideration.

What exactly does the purchasing process look like on TikTok Shop?

TikTok Shop is not a separate app, but a new feature added to the TikTok platform. There’s no need to create a new account or install a new app. This solution provides access to a wide group of potential consumers, as every existing TikTok user over the age of 18 can familiarize themselves with the new feature. This solution gives businesses multiple ways to reach consumers. The platform offers a separate tab, “Shop,” where users can search for specific products using filters and categories, or browse recommended items based on their activity on the platform.

Products offered by sellers using the TikTok Shop service can also be viewed on the “For You Page” tab. This is the subpage most frequently visited by users. This option is especially useful when a company decides to use influencer marketing. A creator posts a video promoting a selected product, and buyers are immediately presented with a purchase button at the bottom of the page. Consumers can also directly access the profiles of brands and creators to find the products they offer or promote.

The latest feature, TikTok Live, is gaining popularity. Before the live stream begins, the seller or influencer adds products available in the TikTok Shop. During the live stream, the host can showcase products, communicate with users, and answer their questions via chat. This can increase the credibility of the product and the seller, as well as encourage consumers to make a purchase, which they can do without interrupting the stream.

The very process of posting ads on TikTok Shop helps build consumer trust. Becoming a seller requires thorough verification, which the TikTok platform conducts to protect users from unreliable and fictitious businesses.

The first step to becoming a seller is to log in to your TikTok Seller Center account using your email address, phone number, or existing TikTok account. You’ll also need to fill out an application form with information that proves your seller credentials, such as your company name, address, and contact information.

After successful verification, the seller completes their store profile, adding a description, name, logo, seller details, addresses, customer service information, and tax information. It’s also necessary to configure payment and delivery methods, including the shipping address, available delivery methods, order processing time, and return policy. Connecting the store dashboard to a regular TikTok account is also crucial. This allows for tagging offered products in live videos, etc. The seller then has the option to publish their product, including the title, description, price, available models, and inventory. The platform also allows businesses to add listings by importing a product catalog from another sales platform.

After a consumer makes a purchase, the seller receives a sale notification in the TikTok Seller Center. The seller is then responsible for packaging and shipping the item to the user, which can be done manually or using external order processing systems.

Distance selling and consumer rights

The TikTok Shop platform offers the option of concluding a sale via a distance contract. This does not require the parties to be physically present at the same time, but rather requires at least one means of distance communication (Act of 30 May 2014 on consumer rights, Article 2). Therefore, when making a purchase through the TikTok Shop, consumer rights are governed by national and European Union law.

In Poland, the primary legal act regulating these activities is the Act of May 30, 2014, on Consumer Rights. Article 12 requires businesses to clearly inform consumers in distance contracts, including the method and deadline for contract execution, the total price including taxes, the right to withdraw from the contract, the complaint procedure, and the seller’s identifying information. The TikTok Shop platform is therefore obligated to provide the required information to the user before finalizing the order via the app. An important regulation is also included in Article 17 of the aforementioned Act and concerns the requirement to design the interface in a way that confirms the consumer’s awareness of the obligation to pay. In the case of platforms that allow order completion via a “button,” it must be clearly marked, e.g., “I buy with an obligation to pay” or “I buy and pay.” Otherwise, the contract is not concluded. The requirements described above are referred to as ” button ” solution ” and are intended to protect consumers from accidentally concluding paid contracts. Alternative obligations also arise from the Directive of the European Parliament and of the Council of 25 October 2011 on consumer rights.

The Consumer Rights Act also implements the EU Commodity Directive (2019/771), introducing uniform standards for the conformity of goods with the contract. A trader is liable for any lack of conformity of goods with the contract upon delivery and for two years from the date the discrepancy is discovered. The Act also governs basic consumer claims in the event of non-conformity, including repair or replacement of the goods, and if this is not possible, a price reduction or withdrawal from the contract.

Given that the sales strategy on the TikTok Shop platform relies on recommendation algorithms and influencer marketing, the Omnibus Directive (EU) 2019/2161 of November 27, 2019, plays a significant role in consumer empowerment. Its regulations introduce the obligation to provide information about the lowest price, disclose whether reviews were published by verified consumers, and indicate whether the seller is a business or an individual. The Omnibus Directive therefore increases consumer awareness and allows consumers to make more rational and manipulation-free purchasing decisions.

Digital Services Act Regulation

Due to their global nature, online platforms reach hundreds of millions of users. Content published through them can reach a very wide audience, thus influencing social, political, and economic relations. Massive social networking sites, therefore, go beyond simply providing entertainment or communication services and digital space, and are beginning to shape the reality around us.

The strong influence of individual platforms on current international relations has initiated more stringent oversight, including through the provisions of Regulation (EU) 2022/2065 of the European Parliament and of the Council of 19 October 2022 on the single market for digital services and amending Directive 2000/31/EC (Digital Services Act – DSA). Based on this regulation, TikTok was recognized by the European Commission as a Very Large Online Platform (VLOP). This status is granted to portals with monthly active users exceeding 10% of the EU population. TikTok exceeded the threshold and was classified as a VLOP in 2023.

Platforms deemed globally influential are not solely responsible for removing illegal content. They are also required to counteract the negative consequences that may arise from their operation. Among the most important responsibilities of very large online platforms (VLOPs) is the identification and assessment of systemic risks and the potential implementation of proportionate mitigation measures. Impacts on fundamental rights, democratic processes, the protection of minors, public safety, and the dissemination of illegal content are particularly scrutinized. Once a specific risk is identified, measures are planned to counteract its escalation, such as modifying the user interface or changing certain advertising practices. In emergency situations, the European Commission has the authority to require VLOP providers to assess the platform’s impact on the development of the crisis, implement specific mitigation measures, and submit periodic reports on the effectiveness of these actions. Very large online platforms are also required to undergo an annual independent audit of compliance with the Digital Services Act (DSA) and regularly publish comprehensive reports on their activities. The aim of this action is to ensure transparency of the platform’s operation towards users and supervisory authorities.

The DSA also regulates advertising by introducing the requirement to maintain a public ad repository. This repository should include, among other things, the ad’s content, advertiser, funding entity, broadcast time, and the number of recipients. This solution is intended to ensure transparency in advertising messages and enable social and scientific analysis of platforms’ promotional activities. Restrictions have also been imposed on recommendation systems. This means that VLOPs are required to provide users with at least one way to display content that is not based on profiling, meaning it does not use user activity history or data. To monitor platforms’ compliance with the EU regulation, it is also possible to impose a requirement to share data on, for example, the performance of recommendation algorithms with the European Commission, national digital service coordinators, or verified researchers.

TikTok, however, is not subject only to the obligations of very large online platforms. It is subject to all regulations provided for in the Digital Services Act. According to Article 26, each advertisement must be clearly identified as promotional material and indicate the advertiser, the funding entity, and the mechanism by which it was tailored to the user. This restriction is particularly useful for the TikTok Shop platform, where sponsored content is commonly created in the manner of regular content published by creators. Limiting the phenomenon of so-called hidden advertising through the provisions described above aims to increase user awareness.

One of the DSA’s key goals is also the protection of minors. When designing their services, platforms are required to consider a high level of protection for minors and their privacy. It is prohibited to display advertisements based on the profiling of minors when the platform has knowledge of the user’s minor status. The goal is to limit the use of children’s data for marketing purposes and reduce the risk of addictive use of the app.

The European Commission has also become concerned about potential negative consumer behavior resulting from the increasing transformation of large social media platforms into e-commerce portals. Complex profiling algorithms, influencer marketing, and instant purchases can encourage users to make impulsive decisions or even become dependent on purchasing processes. Articles 25 and 27 of the Consumer Protection Act (DSA) mitigate this risk. Designing web interfaces that manipulate or complicate consumer decision-making – so-called dark patterns – is prohibited. Examples of unacceptable solutions include hiding options that are less favorable to the business, making it difficult to unsubscribe from services, or designing buttons that encourage a specific choice. Users should also be fully aware of how the recommendation system works; therefore, platforms are required to clearly present its main parameters and the possibility of changing the content suggestion method.

Tamper protection and dark patterns

A key premise of the TikTok Shop platform is the immediacy of purchases. While this solution is very beneficial for businesses and, typically, consumers, it can lead to abuse. Sales without leaving the app, a simplified order completion process, and algorithmic personalization of recommended products seem to provide greater convenience when shopping online. However, some activities can be classified as ” dark patterns”, manipulations used to mislead users and influence their decisions. Because the practices described above can lead to impulsive behavior and distort consumer will, they may be treated as unfair market practices and subject to criminal penalties.

The Act of 23 August 2007 on Counteracting Unfair Market Practices defines an unfair market practice as a sale that is contrary to good practice and significantly distorts or may distort the market behavior of the average consumer before, during or after the conclusion of a product agreement , in particular a misleading market practice and an aggressive market practice (Act of 23 August 2007 on Counteracting Unfair Market Practices, Article 4). The main grounds for considering a market practice misleading include the dissemination of false information or truthful information in a potentially misleading manner. Such misleading information typically concerns the existence of a product, its type or availability, price, the method of price calculation, or the existence of a special price advantage.

To encourage immediate purchases, sellers pressure buyers with messages suggesting limited availability or a limited-time promotion for a specific product. Examples of such messages include phrases like “100 people are viewing the product,” “offer ends in 2 hours,” or “only 4 items left.” This practice is not illegal and is one of the most common marketing mechanisms. Problems arise when the website or portal is programmed to continually extend promotions, the offer doesn’t actually expire after the specified date, or the counter restarts upon page refresh.

Misleading practices, such as suggesting the limited nature of a permanently available offer, and aggressive practices, such as exerting time pressure, may result in legal consequences. In addition to the aforementioned Act of 23 August 2007 on Combating Unfair Commercial Practices, this issue is also regulated by Directive 2005/29/EC concerning unfair business-to-consumer commercial practices in the internal market. This directive distinguishes between misleading commercial practices and aggressive commercial practices. Together, they constitute unfair commercial practices, which include, in particular, actions that are contrary to the requirements of professional diligence and that significantly distort or are likely to significantly distort the economic behavior of the average consumer who reaches or is targeted by the practice, or the average member of a group of consumers if the commercial practice is targeted at a specific group of consumers (Directive 2005/29/EC of the European Parliament and of the Council of 11 May 2005 concerning unfair business-to-consumer commercial practices in the internal market and amending Council Directive 84/450/EEC, Directives 97/7/EC, 98/27/EC and 2002/65/EC of the European Parliament and of the Council and Regulation (EC) No 2006/2004 of the European Parliament and of the Council (“Unfair Commercial Practices Directive”), Chapter 2, Article 5, paragraph 2).

Directive (EU) 2019/2161 of the European Parliament and of the Council of 27 November 2019 (Omnibus Directive) is also intended to combat various manifestations of the “dark patterns” phenomenon. It requires sellers to disclose the lowest price for a product within the last 30 days. This is intended to prevent the practice of artificially inflating prices and then announcing sales. The practice of fake reviews has also been curbed by introducing an obligation to disclose whether and how reviews posted on sales websites are verified. The Omnibus Directive also introduced the requirement to indicate on marketplace platforms whether the seller is a private individual or a professional entrepreneur, so that consumers are aware of who they are buying from.

The European Union has also introduced restrictions related to the design of digital services, contained in Regulation 2022/2065 on the Digital Single Market (DSA). As a result, online platform providers cannot design, organize, or operate their online interfaces in a way that misleads, manipulates, or otherwise interferes with or limits the ability of service users to make free and informed decisions. The regulation therefore directly addresses the fight against “dark patterns,” i.e., website design practices that deliberately use deceptive techniques, such as pre-selected consents or difficult unsubscribes.

The President of the Office of Competition and Consumer Protection (UOKiK) has broad authority to protect consumers from the unfair commercial practices mentioned above. Under the Act of 16 February 2007 on Competition and Consumer Protection, one of his powers is to protect the collective interests of consumers, including through provisions counteracting unfair market practices. If necessary, the UOKiK President may initiate proceedings against a business, ordering it to cease the unfair practice, or requiring the seller to remedy the effects of the violation. Should a business fail to comply with the guidelines, he may impose a fine of up to 10% of the business’s turnover in the previous year.

The number of legal acts, including EU documents, regulating unfair commercial practices reflects the considerable interest in this issue among both legislators and consumer protection authorities. TikTok Shop, as a social commerce model, is not illegal. It utilizes mechanisms combining influencer marketing, personalization, and emotional impact on the recipient, but the design of the user interface is crucial for this platform. The popularity of mass sales portals has contributed to the increasing use of “dark patterns” by businesses over the past few years. For this reason, the European Union and the Office of Competition and Consumer Protection (UOKiK) are increasingly rigorously monitoring sales tactics and issuing new legal acts to protect consumers and their free will when making purchases.

Influencer Marketing and Advertising Law

The effectiveness of influencer marketing stems from combining advertising with the ability to make an immediate purchase. Affiliate links, product tags, or direct purchase buttons, such as those on the TikTok Shop platform, are displayed beneath posts, videos, or other promotional materials. This purchasing model has proven effective by significantly simplifying the ordering process, thus reducing the time consumers spend considering the rationale behind the transaction.

The popularity of the marketing strategy described above stems from its perception by users, who perceive it as authentic and credible. Influencers present promoted products in a natural way, integrating them into their daily routine. However, if the material does not solely reflect the creator’s personal opinion but is created after receiving a benefit in return, it is considered commercial communication. This means it is subject to legal regulations on advertising and consumer protection. In Poland, influencers should clearly label advertising content in accordance with the Recommendations of the President of the Office of Competition and Consumer Protection. These regulations are intended to prevent misleading users.

Only content regarding a product that the influencer purchased independently and for which they did not receive remuneration or other benefits can be marked as a private opinion. Such material contains genuine feelings and opinions and therefore does not constitute advertising under the law and is not subject to advertising law. This is the most credible and reliable form of review for potential consumers, as it was created by a person not under any obligation to the manufacturer.

A manufacturer may enter into an agreement with an influencer to promote a product in exchange for a free product, financial benefit, or other form of remuneration. This creates legally regulated advertising. It may take the form of a post, report, or live broadcast in which the creator demonstrates how they use the product and its positive properties. Due to the natural presentation of the product as an everyday element, the recipient may have difficulty distinguishing a genuine recommendation from commercial content. The Act of August 23, 2007, on Counteracting Unfair Market Practices, classifies the act of concealing a promotional message as a misleading omission. Failure to clearly indicate the commercial nature of the material may hinder consumers’ proper assessment of the message and directly influence their purchasing decisions.

Another common advertising strategy is to feature a product integrated into published content without directly promoting it, for example, by placing it in the background of the material. This phenomenon is called product placement. Activities covered by advertising and consumer protection law also include, among others, affiliate and partner links, ambassador programs, and partner competitions. In Poland, these practices must contain clear, understandable to the average recipient, and visible advertising labels from the very beginning, such as “advertisement,” “paid collaboration,” or “sponsored content.” The Office of Competition and Consumer Protection (UOKiK) also recommends the use of two-level labeling, meaning that, in addition to the information contained in the content, the platform’s functionality must also be used to announce the paid collaboration. Detailed guidelines can be found in the Recommendations of the President of the UOKiK regarding the labeling of advertising content by influencers. Material is considered advertising content not only when the influencer receives monetary compensation in exchange for its creation. The same obligation applies when promoting your own business, receiving a free product or service, or obtaining a sales commission via an affiliate link or discount code (Recommendations of the President of the Office of Competition and Consumer Protection regarding the marking of advertising content by influencers).

In the event of non-compliance with the Recommendations of the President of the Office of Competition and Consumer Protection regarding the labeling of advertising content by influencers, pursuant to the Act of 16 February 2007 on Competition and Consumer Protection, the Office of Competition and Consumer Protection (UOKiK) conducts proceedings against entrepreneurs using practices that violate the collective interests of consumers. Actions may be taken against advertisers, influencers, and marketing agencies. Therefore, responsibility for incorrect labeling of advertising content rests not only with the creator publishing the material but also with all entities participating in organizing the promotional campaign. One of the sanctions that the President of the UOKiK has the right to impose is a financial penalty. Incorrectly labeled promotional material can also be considered surreptitious advertising. Due to the dynamic development of influencer marketing, the proper creation of marketing content is currently widely subject to UOKiK scrutiny. Therefore, it is worth clearly and understandably labeling sponsored publications, among other things, to avoid significant financial penalties.

Personal data protection

TikTok Shop, as a hybrid social network and e-commerce platform, processes a significant amount of data related to both user activity and purchasing processes. The app’s operation is based on audience profiling and matching the most relevant content. Therefore, the platform’s operations are subject to the provisions of Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data (GDPR).

In addition to user data such as name, surname, contact details, shipping addresses, and payment information, media platforms also collect information that allows for behavioral analysis. Time spent browsing specific products, interactions with ads, and the history of items added to carts or wish lists allow the TikTok Shop platform to create a personalized recommendation system based on past activity. This phenomenon creates so-called behavioral advertising, a marketing strategy based on user profiling using advanced algorithms to predict future purchasing decisions. The concept of profiling refers to the automated processing of personal data, particularly for the purpose of predicting a user’s economic situation, personal preferences, interests, health, and location.

According to the GDPR, profiling is permissible, but it also comes with a number of obligations. Platforms are obligated to transparently inform users about, among other things, the purposes of profiling, the legal basis for data processing, the consequences of the actions taken, and their rights, including the right to object to profiling. Data of minors is particularly protected. Due to the growing popularity of the TikTok app among young users, it was necessary to subject it to special regulations in this regard. In the area of information society services, the processing of data of children over 16 years of age is lawful. An exception is made for situations in which a person with parental authority or guardianship provides prior consent. However, EU member states may introduce a lower age limit in their laws, but it must be at least 13 years old, as is the case in Poland, for example. To ensure that platforms enforce their obligations related to the protection of minors, they should use appropriate age verification mechanisms. In practice, however, this solution requires further improvement due to the common practice of users providing false data during registration.

The President of the Office of Competition and Consumer Protection (UOKiK) plays a crucial role in protecting users, especially the collective interests of consumers. He is authorized to take action against entrepreneurs who engage in unfair market practices, design manipulative interfaces, and so on. Personal data protection, however, falls primarily within the remit of the Office for Personal Data Protection (UODO), which oversees compliance with the GDPR and the secure processing of information by companies and institutions. Due to its global influence, TikTok has attracted increasing attention from EU authorities in recent years and is becoming the subject of more frequent inspections. Due to the platform’s European headquarters being located in Ireland, the relevant supervisory authority is the Irish Data Protection Commission (DPC). For example, in 2025, this institution imposed a fine of €530 million on ByteDance, the app’s owner. The fine was imposed on the transfer of user data from the European Economic Area to China in violation of the GDPR and the failure to demonstrate data protection at the level guaranteed in the EU.

The GDPR is supplemented by Directive 2002/58/EC of the European Parliament and of the Council of 12 July 2002 concerning the processing of personal data and the protection of privacy in the electronic communications sector (Directive on privacy and electronic communications), which regulates, in addition to the processing of personal data, the confidentiality of electronic communications, also known as ePrivacy. Due to the scope of its regulations, the provisions of this directive have particular relevance to the TikTok Shop application. The platform uses numerous tracking technologies, such as cookies and mobile device advertising identifiers, to monitor user activity. Information may be stored on a user’s device or accessed only after obtaining prior consent. Exceptions are made only for technologies strictly necessary to provide the service requested by the user, such as remembering a shopping cart. An additional ePrivacy regulation was also envisaged, the purpose of which was to replace the current directive and harmonize the personal data protection rules applicable in all EU Member States. The changes were to include, among other things, simplifying the rules regarding cookies. However, the project encountered legislative difficulties and was not adopted by decision of the European Commission.

Abuse of Market Power and the Digital Markets Plan

The dynamic expansion of the largest digital platforms’ influence has led to the need to adapt competition law to the new situation, particularly in the digital market. To this end, the European Union adopted Regulation (EU) 2022/1925 of the European Parliament and of the Council of 14 September 2022 on contestable and fair markets in the digital sector and amending Directives (EU) 2019/1937 and (EU) 2020/1828 (Digital Markets Act – DMA). The Act introduces the concept of a gatekeeper. This term refers to an entrepreneur with an influential position in the internal market, through which business users reach end users. A gatekeeper provides a core platform service and maintains an established market position.

The dominant position of a gatekeeper is also associated with a number of obligations. Among other things, the practice of self-preferencing, which involves favoring one’s own products or services over the offers of other businesses using the platform, is prohibited. In the case of TikTok Shop, this could involve using recommendation algorithms to increase the visibility of products promoted by individual sellers, without applying objective and fair advertising criteria. This type of favoritism and limiting the reach of individual entities could lead to a distortion of fair competition between businesses using TikTok Shop for sales purposes.

By decision of the European Commission, BytaDance Ltd. was granted gatekeeper status solely for the operation of the TikTok application as a social media platform. The DMA regulations governing the gatekeeper position do not apply directly to TikTok Shop, but they may impact the rules for recommending products and using entrepreneurs’ data.

Media law and audiovisual regulations

Audiovisual materials are the primary tool for promoting and selling products on the TikTok Shop marketplace. Therefore, the app’s operations are also subject to scrutiny for compliance with media law and regulations governing audiovisual media services. The dominant role in this regard is played by Directive 2010/13/U of the European Parliament and of the Council of 10 March 2010 on the coordination of certain provisions laid down by law, regulation, or administrative action in Member States concerning the provision of audiovisual media services (Audiovisual Media Services Directive – AVMSD) and the Broadcasting Act of 29 December 1992, which implements it into Polish law. As a result of the amendment to the Act of 11 August 2021, the regulations have been extended to video-sharing platforms, including the TikTok app.

Video-sharing platforms are primarily obligated to implement appropriate measures to protect minors from harmful content that could negatively impact their moral, mental, or physical development. These provisions have been implemented into Polish law through Article 47e of the Broadcasting Act, which mandates, among other things, the marking of potentially inappropriate content with special graphics for young viewers. These regulations are particularly important for the TikTok Shop platform due to the constantly growing number of underage users. Posting content that spreads hatred and discrimination is also prohibited.

TikTok Shop, a hybrid social media platform and e-commerce platform, is often used to publish so-called audiovisual commercial communications—images used to directly or indirectly promote goods, services, or individuals (Directive 2010/13/U of the European Parliament and of the Council of 10 March 2010 on the coordination of certain provisions laid down by law, regulation or administrative action in Member States concerning the provision of audiovisual media services, Article 1). Article 9 of the AVMSD requires member states to ensure that such communications are easily recognizable, thus prohibiting hidden audiovisual commercial communications. The use of subliminal techniques or the inclusion of discriminatory content would also be illegal. The National Broadcasting Council (KRRiT) is responsible for ensuring compliance with audiovisual law. Its remit includes, among other things, overseeing the activities of video-sharing platform providers.

The sales method used by TikTok Shop may seem analogous to teleshopping, offerings directly to consumers to deliver goods or services in exchange for payment. This modern form of interactive audiovisual commerce (live shopping) bears numerous similarities to traditional teleshopping. The mechanisms of both aforementioned sales methods involve presenting the product, its specific features, available options, and generally encouraging the recipient to purchase. However, teleshopping is targeted at a general, anonymous audience who may only be interested in the recommended product. Meanwhile, TikTok Shop relies on advanced algorithms that target promotional content to users who, based on their previous activity, have shown interest in similar content.

Platform liability under e-commerce regulations

The original act regulating the legal liability of online platforms in the European Union was Directive 2000/31/EC of the European Parliament and of the Council of 8 June 2000 on certain legal aspects of information society services, in particular electronic commerce, in the Internal Market (Directive on electronic commerce). Its foundation was the so-called safe harbor principle, i.e., the principle of limited liability of online intermediaries. According to this principle, hosting service providers and online platforms were not liable for content or goods published by users, provided they had no actual knowledge of the illegal nature of the content or goods or services or, upon obtaining such knowledge, promptly removed any infringements. Furthermore, the directive did not impose a general obligation on platforms to monitor content published by users.

However, the ongoing development of digital platforms has made it necessary to amend the current liability model. Regulation (EU) 2022/2065 – Digital Services Act (DSA) – came into effect on February 17, 2024. This regulation does not eliminate the principle of limited liability but significantly expands monitoring obligations, especially for very large online platforms (VLOPs). One of the key obligations introduced under the new regulations is the Know Your Business Customer (KYBC) principle. This regulation aims to increase the safety of consumers shopping online by limiting sales conducted by dishonest or anonymous traders. Before enabling sales through its platform, an online platform must collect and verify basic data identifying the seller. The required information includes, among others, the trader’s name, registered office address, contact details, registration number in the relevant register of traders, and the trader’s payment account details. In the event of refusal to provide the specified data or providing it falsely, the platform should prevent the trader from conducting sales until the situation is resolved.

A problematic issue related to the TikTok Shop app is defining the platform’s responsibility for transactions conducted by sellers using it. Although TikTok Shop formally acts as an online intermediary, it can be argued that its operating mechanism goes beyond passive hosting. A recommendation system using algorithms, promoting offers, and providing marketing and analytical tools to sellers are the mechanisms TikTok Shop uses to shape consumer behavior and purchasing decisions. The platform’s influence on the visibility of offers and the order fulfillment process may support assigning it broader responsibilities in overseeing the online sales process.

Regulations on electronic communications, including the European Electronic Communications Code and the Polish Electronic Communications Law

The TikTok Shop platform does not constitute an electronic communications service under European Union law, but its operations provide for various forms of electronic communication. TikTok Shop’s use of push notifications, in-app messages, and marketing communications requires the platform to comply with regulations governing electronic marketing and the protection of user privacy in electronic communications. The primary legal acts regulating these aspects are Directive (EU) 2018/1972 of the European Parliament and of the Council of 11 December 2018 establishing the European Electronic Communications Code (EECC) and the Act of 12 July 2024 – Electronic Communications Law.

The primary function of TikTok Shop is to enable entities to sell goods through the social media platform. Article 2 of the European Electronic Communications Code defines an electronic communications service as the transmission of signal transmissions or the provision of interpersonal communications services. The mere ability to exchange messages between users or with sellers does not automatically qualify the TikTok Shop platform as a provider of electronic communications services, as this is not its core competency and does not constitute its core business. However, because electronic communications are primarily used for marketing purposes, it is obligated to comply with regulations governing direct marketing and the protection of user privacy.

Push notifications, messages sent directly to users’ mobile devices, are an increasingly popular marketing solution. TikTok Shop uses them to provide information about order status, discounts, time-limited campaigns, or the launch of live shopping. Transactional notifications regarding order fulfillment, shipping, or payment status are typically part of the contract and do not require marketing consent. However, notifications encouraging potential consumers to make a purchase are classified as direct marketing and, in accordance with electronic communications law, require prior user consent.

The practice of using automated calling systems and electronic means of communication for advertising purposes without the user’s prior consent is also prohibited. Users should be clearly informed about the purpose of receiving marketing communications, the data controller, and the possibility of withdrawing consent, which should not result in any negative consequences. With respect to the TikTok Shop platform, the above position means that it is unlawful to send promotional content to users solely based on the fact that they have an account on the app.

TikTok Shop is the clearest example of how thin the line between entertainment, advertising, and commerce has become – a one-tap purchase woven into a stream of content is now as effortless as liking a video. Yet that convenience comes at a price: the closed-loop model and algorithmic personalization shrink the time left for rational reflection, while responsibility for protecting the consumer shifts increasingly away from the buyer and onto the platform and the legislator. EU and national regulations – from consumer law, through the DSA and DMA, data protection and safeguards for minors, all the way to media and electronic communications law – form a web meant to counterbalance the platform’s power and restore the buyer’s awareness of their own choices. TikTok Shop thus remains a dual phenomenon: on one hand a groundbreaking innovation in digital commerce, on the other a test of whether the law can keep pace with a technology that sells faster than we can think.

Artykuł A Store in Your Pocket, the Law in the Background: TikTok Shop Under the Regulators’ Lens pochodzi z serwisu KIELTYKA GLADKOWSKI LEGAL | CROSS BORDER POLISH LAW FIRM RANKED IN THE LEGAL 500 EMEA SINCE 2019.

The phenomenon of fake reviews in the digital space has evolved from a marginal image issue to a central focus of market supervision authorities and EU legislators. The contemporary ontology of this phenomenon extends beyond primitive content fabrication to encompass any form of communication that, by distorting the actual consumer experience, misleads the recipient, directly influencing their decision-making process. Legally, a fake review is considered not only a completely false message, but also one that, by omitting important facts or manipulating context, creates a false impression of the quality of a product or the reliability of a seller. This practice is classified as unfair commercial activity if its nature causes or is likely to cause the average consumer to make a transactional decision they would not otherwise make, thus violating the fundamental principles of fair dealing.

The typology of activities considered unfair rests on several fundamental pillars, the most blatant of which is direct fabrication. This involves posting or commissioning the creation of false recommendations from specialized external entities, such as marketing agencies, which directly violates regulations on combating unfair market practices. Another mechanism is selective manipulation, in which a business intentionally manages the visibility of reviews by removing, concealing, or delaying the publication of negative reviews while favoring positive ones. Such action distorts the image of actual customer satisfaction and is considered misleading regarding the essential characteristics of a product or service. An equally significant aspect is feigned verification, i.e., declaring that reviews come from real buyers without implementing proportionate and reasonable steps to verify their authenticity, which constitutes a direct violation of the disclosure obligations imposed by the Omnibus Directive.

Contemporary market practices have also evolved more subtle forms of manipulation, such as astroturfing, which involves creating artificial social support through employees or store owners posing as independent consumers. These activities often involve the manipulation of user profiles, where images generated by artificial intelligence algorithms are used to authenticate fictitious accounts, creating false social proof. Each of these practices, regardless of their technological sophistication, is subject to strict scrutiny by competition and consumer protection authorities.

The role of the President of the Office of Competition and Consumer Protection and the responsibility of management boards

The President of the Polish Office of Competition and Consumer Protection (UOKiK) serves as a central regulator in the legal system, endowed with rigorous powers to counteract violations of collective consumer interests. The main disciplinary instrument at the authority’s disposal is an administrative fine, which can be imposed in the amount of 10% of the turnover achieved by the entrepreneur in the financial year preceding the year of issuance of the decision. The amount of the fine is not determined arbitrarily, but rather results from precisely defined criteria, which include, above all, the scale of the violation, its duration, and the degree of intentionality of the perpetrator. Importantly, this fine is intended to serve not only a repressive function but, above all, a preventive and deterrent one, discouraging other market participants from engaging in similar unfair practices involving the manipulation of reviews or misleading as to the authenticity of reviews.

The enforcement procedure in consumer matters is designed to ensure high effectiveness of supervisory activities. A business subject to a sanction is obligated to settle the fine within 14 days of the decision becoming final, which directly contributes to the state budget. A crucial procedural element is the prejudicial nature of the decisions of the President of the Office of Competition and Consumer Protection (UOKiK), which means that the authority’s findings regarding violations of the law are binding on common courts in compensation cases brought by injured customers. This legal structure significantly facilitates consumers in pursuing civil claims, as they do not have to prove the illegality of the store’s actions, focusing solely on demonstrating the damage suffered. The office’s activity in recent years, reflected in numerous proceedings against e-commerce leaders, confirms that protecting the transparency of reviews has become a regulatory priority, translating into real and severe financial consequences for violators.

The contemporary model of liability in consumer protection law departs from a concept focused solely on the business entity, shifting the burden of sanctions also to individuals who actually manage the enterprise. The President of the Office of Competition and Consumer Protection (UOKiK) has the authority to impose a personal fine of up to PLN 2,000,000 on a manager. This liability is triggered by demonstrating that the manager has intentionally allowed – through their actions or conscious omissions – the company to violate collective consumer interests. In case law, the degree of management involvement in decision-making processes regarding marketing and communications is crucial. This liability may therefore affect a management board member who approves a budget for obtaining reviews from external opinion farms or ignores the lack of implementation of verification procedures under the Omnibus Directive, despite being aware of such deficiencies.

It should be emphasized that the responsibility of managers is autonomous and independent of any penalty imposed directly on the entrepreneur. This is intended to provide a strong incentive for management to build internal compliance structures and actively oversee the entity’s operational ethics. In the era of digitalization of trade, where algorithms and automation of marketing processes can generate violations on a massive scale, the personal financial risk of managers is intended to compel prioritizing compliance as the foundation of business strategy. Therefore, the systemic fight against false reviews is implemented not only through sanctions against corporate structures but also by disciplining those who actually shape companies’ market policies. This, according to the legislature, is intended to ensure long-term improvement in integrity standards in electronic trading.

The Omnibus Directive and the blacklist of market practices

The implementation of the Omnibus Directive into the Polish legal system significantly redefined transparency standards in e-commerce, introducing mechanisms that directly address the systemic manipulation of consumer reviews. A key instrument in this regard is the so-called blacklist of market practices, which constitutes a catalog of behaviors considered unfair in all circumstances, eliminating the need for supervisory authorities to conduct a case-by-case analysis of the consequences of a given action. Classifying these market torts as unfair practices aims to eliminate evidentiary difficulties, as their mere existence exaggerates the entrepreneur’s wrongdoing. This legal framework not only strengthens the consumer’s position but, above all, simplifies the evidentiary process, making the fight against e-commerce abuse more effective and predictable for market participants. The foundation of the new regulations is an absolute prohibition on manipulating the verification and authenticity of product recommendations, which imposes an active obligation on sellers to implement procedures to verify the origin of reviews.

Under the current wording of the regulations, it is considered an unfair market practice for a trader to claim that product reviews were posted by consumers who actually used or purchased the product, in situations where reasonable and proportionate steps were not taken to verify their authenticity. This practice violates the consumer’s right to reliable information, which is essential for making an informed decision about purchasing the product, and violating it constitutes conduct contrary to good practice. The law prohibits not only posting completely false reviews, but also commissioning third parties to create them, or transferring recommendations between products with different parameters, which is referred to as review hijacking. Other offenses listed in the catalog are treated equally severely, such as using false quality certificates without appropriate authorization or using surreptitious advertising, which involves using editorial content to promote a product without clearly identifying the paid nature of the communication. Aggressive techniques are also considered particularly burdensome, including mass spamming and forced selling, which involves demanding payment for products delivered to the consumer without their prior order.

The blacklist also eliminates techniques such as bait advertising and direct persuasion of children to purchase, which aims to protect the integrity of the consumer decision-making process from manipulation. This protection of minors stems from their particular vulnerability to advertising messages and their inability to critically assess the persuasive nature of commercial offers. Expanding the list to include a ban on posting or commissioning another person to post false reviews for the purpose of promoting products significantly complements the system, preventing brands from using agencies that fabricate social evidence. It is emphasized that any form of distortion of the actual image of a product’s popularity constitutes a violation of the collective interests of consumers, which entitles the President of the Office of Competition and Consumer Protection (UOKiK) to intervene under public law as soon as a threat to the interests of all market users arises.

A particularly significant and painful consequence of these unfair techniques for entrepreneurs is a specific civil law sanction in the form of an extended right of withdrawal from the contract. If an e-store engages in practices listed in the prohibited catalog or fails to comply with information obligations regarding review verification, the statutory return period granted to the buyers is extended from 14 days to a full 12 months. This mechanism is a direct consequence of the assumption that, in the absence of reliable information, the consumer could not have expressed a fully informed intention to purchase, which suspends the running of standard mandatory deadlines. Systematic combating of review fraud and the use of black market practices is therefore becoming not only a matter of business ethics but the foundation of legal security and stability for every entity operating in the e-commerce sector. Neglect in transparency can lead to mass claims for refunds, posing a real threat to the operational liquidity of the company.

Manipulation Architecture and Platform Obligations under the Digital Services Act (DSA)

The phenomenon known as dark patterns constitutes a sophisticated form of interference in the user’s decision-making process, based on the deliberate use of interface architecture to distort their autonomy of will. Manipulative design patterns are not merely a manifestation of aggressive marketing, but a systematic designer’s action aimed at inducing a specific cognitive bias in the consumer, which ultimately leads to a purchase decision they would not have made in conditions of full transparency. The psychological foundation of these actions is the use of heuristics, i.e., simplified rules of reasoning and automatic thinking, which in the fast-paced environment of e-commerce transactions make the user susceptible to subliminal suggestions. This phenomenon has evolved from simple forms of persuasion to advanced interface manipulation, where the line between inducement and fraud is deliberately blurred to maximize conversion at the expense of the interests of the weaker party in the legal relationship.

A particularly significant area of application of these practices is the system for presenting reviews and suggesting their authenticity, where manipulation takes the form of so-called interface interference. Businesses often employ patterns involving selective content display, which in practice means deliberately hiding negative reviews on subsequent pages of the website while simultaneously highlighting only enthusiastic reviews on the product’s home page. This practice violates the model of the average consumer, who has the right to expect that the image presented of a product’s popularity and quality is reliable and has not been subjected to arbitrary filtering. Manipulation in the sphere of social evidence also includes fabricating popularity indicators, such as false messages about the number of people viewing a given product at a given time or false offer duration counters, which create an artificial sense of scarcity in the user and pressure them to immediately close the transaction. Under the Polish Act on Combating Unfair Market Practices, these activities may be classified as misleading because they distort the actual market conditions, preventing a rational comparison of offers.

Another dimension of manipulation is the technique known as confirmation shaming, which in the sphere of opinion writing involves the use of evaluative and emotional language to coerce users into specific behaviors, for example, through unsubscribe buttons suggesting a lack of consumer awareness. These practices are closely related to the “roach motel model”, where the process of issuing a favorable review is simplified to the maximum extent, while editing, reporting an error, or deleting content requires navigating a complex subpage structure, which is intended to discourage users from correcting false information. In the legal context, such procedural barriers are considered burdensome impediments that violate good practice and the principle of commercial fairness. An analysis of case law and the positions of supervisory authorities indicates that an interface that deliberately hinders users from exercising their rights or changing their minds loses its neutrality and becomes a tool for harming consumer interests.

A fundamental change in the regulatory sphere was brought about by the entry into force of the EU Digital Services Act (DSA), which, in Article 25, explicitly prohibits online platform providers from designing, organizing, and operating interfaces in a way that misleads or manipulates service users. This regulation is overarching and complements the existing consumer protection framework by introducing a direct obligation to maintain neutrality in choice architecture and prohibiting structures that significantly impede users’ ability to make free and informed decisions. Violation of this prohibition entails not only civil law risks but also severe administrative sanctions, which can amount to a significant percentage of the business’s global turnover.

In the sphere of law enforcement, the key role is played by the model design of the average consumer, who is observant and cautious but lacks specialized knowledge of the psychological mechanisms used in interface design. This protection is preventative and abstract in nature, meaning the President of the Office of Competition and Consumer Protection (UOKiK) can intervene in situations where the mere existence of a manipulative pattern poses a real risk of distorting market behavior, without having to wait for measurable financial damage to a specific individual. Effectively combating dark patterns requires businesses not only to comply with the law but, above all, to shift to a design model focused on reliability, where all product information, including opinions, is presented free from coercive mechanisms. Ultimately, interface transparency is becoming a prerequisite for maintaining trust in the digital economy, and the use of sophisticated forms of manipulation is perceived as highly harmful to society, subject to strict assessment in light of the principles of social coexistence.

New obligations for marketplaces regarding moderation and transparency

The entry into force of Regulation 2022/2065, known as the Digital Services Act (DSA), represents a fundamental shift in the liability paradigm for intermediary service providers, particularly marketplaces. This regulation shifts the emphasis from passive content hosting to active oversight of the transparency and security of the digital system, introducing rigorous operational standards aimed at eliminating illegal content while respecting users’ fundamental rights. A key pillar of this reform is the formalization of moderation processes, which until now were often subject to arbitrary internal platform decisions and are now subject to strict procedural rigors contained in the notice-and-action mechanism. Under the DSA, each platform is required to provide easily accessible and user-friendly tools for identifying potentially illegal content, including fake reviews or infringing offers. The mere receipt of a report obliges the provider to promptly and objectively address it.

The evolution of moderation obligations is inextricably linked to the requirement for transparency in decisions, which is achieved through the justification mechanism provided for in the EU regulation. When a marketplace decides to remove content, limit its visibility, or suspend a user’s account, the user is absolutely obligated to provide clear and specific reasons for such action, which is intended to prevent abuse by blocking reliable reviews that are unfavorable to the seller. This system is complemented by a mandatory internal complaint handling system, which allows users to appeal moderation decisions free of charge within a period of at least six months. This constitutes an important procedural guarantee and allows for the correction of potential algorithmic errors. It is indicated that such a legal framework is necessary to counteract the fragmentation of consumer protection, which previously relied primarily on general national clauses that were unsuitable for the scale of operations of global digital entities.

A significant innovation introduced specifically for trading platforms is the “Know Your Business Customer” (KYBC) principle, regulated in the chapter on marketplace transparency. These entities are charged with collecting and verifying information about traders offering their products through their interfaces, including registration data, payment account numbers, and declarations of commitment to offer goods in compliance with EU law. This mechanism aims to eliminate the phenomenon of anonymous sellers, who often promote defective products using fabricated reviews and, after raising capital, disappear from the market, avoiding legal liability. The platform is obligated to suspend services for sellers who fail to submit the required documents, making the marketplace an active guardian of the legality of trade, rather than merely a passive intermediary in trade.

The scope of transparency obligations extends beyond relationships with individual users to include public reporting through the periodic publication of transparency reports. These documents must include detailed data on the number of orders received from national authorities, statistics on content moderation initiated by the platform itself, and information on the use of automated tools in verification processes. For very large online platforms, these rigors are even stricter, including the obligation to conduct annual audits and systemic risk assessments, including analysis of the interface’s vulnerability to manipulation that could negatively impact public safety or consumer protection. The systemic fight against disinformation and unfair market practices is therefore anchored in the full transparency of operational processes, which allows supervisory authorities to continuously monitor the effectiveness of implemented security measures.

Supervision of compliance with these obligations is based on a new institutional architecture, in which national digital services coordinators, working closely with the European Commission, play a central role. The enforcement system for the adopted regulations is based on fines of up to 6% of a provider’s global turnover, which compels compliance with specific cybersecurity standards. This control system is designed to ensure that marketplaces not only implement the required procedures but also apply them reliably and uniformly across the European Union, which is crucial for building consumer confidence in cross-border trade. The introduction of these standards ends the phase of full regulatory freedom for platforms, imposing on them real responsibility for shaping the environment in which the modern exchange of goods and services takes place.

Technological verification mechanisms and modern operating models

Authenticity Suggestion and Pressure Mechanisms

The evolution of digital market oversight has led to the development of mechanisms in which traditional legal instruments are increasingly being replaced by algorithmic jurisdictions based on advanced artificial intelligence systems. The phenomenon known as AI exclusion is a modern form of sanction that, for e-commerce entities, can prove more severe than traditional financial penalties imposed by administrative bodies. The foundation of this process is the integration of data on the credibility of reviews directly with positioning parameters in ranking systems, which means that transparency is no longer merely an ethical obligation but a condition for the technical visibility of an offer. Recommendation algorithms operating within platforms such as Google and Amazon constantly analyze behavioral and linguistic patterns to identify anomalies suggesting manipulation of social evidence. These systems are currently capable of recognizing the structure of texts generated by LLM language models, which are characterized by a specific repetition of phrases and a lack of emotional details typical of authentic consumer experiences. An additional risk factor subject to automatic verification is the so-called review growth rate, where a sudden jump in the number of positive ratings without correlation with actual website traffic or sales volume is interpreted by AI as a warning signal initiating restrictive procedures.

The consequences of an online store being classified by AI systems as posing a high risk of manipulation are immediate and often irreversible in the short term. This mechanism, known in market practice as shadow banning or de-indexing, leads to a drastic decline in visibility in search results and the blocking of offers in advertising systems, effectively cutting the entrepreneur off from key customer acquisition channels. Under the provisions of the Digital Services Act, providers of very large online platforms are required to maintain particular transparency regarding the parameters used in recommendation systems. Article 27 of the aforementioned regulation requires platforms to clearly define in their regulations the key parameters determining information ranking, which aims to limit algorithmic arbitrage and enable entrepreneurs to understand the reasons for a potential decline in their market exposure. It is worth noting that modern risk assessment systems may be classified as high-risk systems within the meaning of the Artificial Intelligence Regulation, which imposes strict requirements on their creators regarding human oversight and the prevention of algorithmic discrimination.

In parallel to restrictive systems, a paradigm known as agentic commerce is developing, in which purchasing processes are carried out by autonomous AI assistants acting directly on behalf of the consumer. In this model, traditional product reviews cease to serve as persuasive texts for humans and become raw input data for machines that filter the market in search of offers with the highest level of verified trust. A key element of this new commerce architecture is the so-called trust layer, built on protocols such as the Universal Commerce Protocol promoted by Google or the Agentic Commerce Protocol developed by OpenAI. These systems are guided not only by price or availability of goods but above all by the certified credibility of the seller’s data, automatically rejecting offers from entities that lack a clear digital traceability of their recommendations. The collaboration of AI assistants with secure payment systems, such as the Agent Payments Protocol, creates a closed ecosystem in which offers at risk of manipulation are excluded at the initial algorithmic selection stage, before they are even presented to the user.

In the era of agent-based commerce, the role of modern shopping assistants is becoming dominant, forcing businesses to redefine their credibility-building strategies. The Context Protocol model and other open-source solutions enable the exchange of context between various AI models and commerce systems, allowing information about unfair practices by a single store to be instantly shared across the entire assistant network. The doctrine suggests that this systematic approach to eliminating abuse is a natural response to the technological ease of fabricating content online. For an e-commerce store, losing its trustworthy status in the eyes of Google or OpenAI algorithms means the modern equivalent of server shutdown, as AI assistants, protecting the interests of their users, will systematically bypass offers that generate manipulative signals. Thus, the fight for authenticity is no longer a mere compliance issue but an existential foundation in the new, automated e-commerce environment, where barriers to entry into the trust layer are becoming increasingly difficult for entities employing pressure mechanisms and suggesting false authenticity.

Compliance as a Service and the Digital Feedback Path

The rapid evolution of the e-commerce market and the increasing professionalization of unfair market practices have forced entrepreneurs to abandon a reactive reputation management model in favor of proactively building a digital immune system. The scale of the challenge facing modern e-commerce is illustrated by analyses of the systematic erosion of trust in the digital sector, pointing to the prevalence of fake reviews and consumer concerns about the mass implementation of generative artificial intelligence for opinion fabrication. This state of affairs creates decision paralysis, where an overabundance of unreliable information, instead of supporting the purchasing process, becomes an insurmountable barrier.

The economic impact of the lack of reliable content verification is directly measurable and translates into tangible operational losses for businesses. The literature emphasizes that exposure to manipulated reviews drastically reduces purchase intentions and brand trust, generating measurable financial losses. The information vacuum filled with false enthusiasm also leads to a phenomenon known as post-purchase dissonance, in which a product that fails to meet expectations is returned to the seller as a complaint or contract withdrawal. Consequently, the lack of investment in transparent review processes generates hidden logistical and operational costs that, in the long run, may outweigh the gains achieved through the temporary increase in conversions driven by manipulation.

In response to increasing regulatory rigor, including the Omnibus Directive, the Digital Services Act (DSA), and the AI Act framework, an operational model known as Compliance as a Service (CaaS) has emerged in market practice. It involves fully outsourcing compliance processes to specialized technology providers who take over the burden of monitoring and verifying content in accordance with current regulations. CaaS allows for the automation of data oversight, which is essential in an environment where the volume of incoming reviews precludes manual oversight without risking accusations of disproportionality. In this approach, compliance ceases to be merely an administrative cost and becomes a component of a strategy for building brand value by guaranteeing the authenticity of every customer touchpoint.

The foundation of the Compliance as a Service model is the maintenance of clean data and the generation of an indisputable digital trace of the review’s provenance. Every published review should be accompanied by a log containing metadata regarding the specific transaction, a unique order number, and delivery status, creating auditable proof of authenticity that can be presented during inspections by supervisory authorities such as the President of the Office of Competition and Consumer Protection. This digital reconstruction of the review process provides the most effective legal shield for businesses, eliminating the risk of allegations of unfair market practices. In the era of algorithmic jurisdiction, where ranking systems favor content supported by digital evidence, having a certified trace of data provenance is becoming a prerequisite for maintaining the market visibility of an offer.

Parallel to technical verification, modern review management systems integrate mediation mechanisms that allow for the amicable resolution of disputes before they are publicly expressed. Market experience suggests that implementing structured review processes allows for the amicable resolution of a significant portion of consumer disputes, effectively preventing the publication of negative reviews resulting from logistical errors. This approach aligns with the principles of reliability and good market practices, building customer relationships based on dialogue rather than solely on the one-way transmission of ratings.

Transaction verification is now becoming the market standard, replacing open, abuse-prone review sections with a system of unique invitations sent only after a purchase is completed. The literature emphasizes that restricting the review process to those who actually purchased the product is the simplest and most effective way to comply with the obligations imposed by the Omnibus Directive. This not only minimizes the risk of severe financial penalties, but above all, provides AI shopping assistants with reliable input data, which, in the new agent-based commerce paradigm, will determine the viability of each entity in the e-commerce ecosystem.

Artykuł Faking reviews in e-commerce – analysis of new legal regulations, algorithmic mechanisms and market practices in the e-commerce sector pochodzi z serwisu KIELTYKA GLADKOWSKI LEGAL | CROSS BORDER POLISH LAW FIRM RANKED IN THE LEGAL 500 EMEA SINCE 2019.

Just a few years ago, online store owners primarily had to ensure terms and conditions, privacy policies, and efficient order processing. Today, this is clearly not enough. EU regulations such as the Omnibus Directive and the Digital Services Act (DSA), as well as the increasing role of artificial intelligence in assessing store credibility, force businesses to consider their platforms much more broadly. It is no longer just about regulatory compliance, but also about building digital trust, which influences a store’s visibility, legal security, and customer purchasing decisions. Below, we present a practical checklist of the most important actions to implement to reduce the risk of sanctions and increase the credibility of an online store.

Practical guidelines for online store owners

I.            Avoiding UOKiK fines and compliance with the Omnibus Directive

a. Implement transactional verification: You should configure your feedback system so that each review you post is technically linked to the unique order number and email address of the customer who actually completed the purchase.

b. Updating the content of the regulations: In the “Rules for publishing opinions” section, the verification procedure should be described in detail, whether all opinions (including critical ones) are published and how the average product rating is calculated.

c. Transparent labeling: Each review should have a clear status indication (e.g., “Purchase confirmed”). If a benefit is provided in exchange for reviews (e.g., a discount code), this information must be clearly and prominently displayed within the review text.

d. Lowest price mechanism: In accordance with the requirements of price transparency, each discount must display the lowest price of the product that was valid in the 30 days prior to the introduction of the discount.

Legal basis: Act of 30 May 2014 on consumer rights ( Journal of Laws of 2024, item 1796, as amended); Directive (EU) 2019/2161 of the European Parliament and of the Council of 27 November 2019 amending Council Directive 93/13/EEC and Directives 98/6/EC, 2005/29/EC and 2011/83/EU of the European Parliament and of the Council as regards the better enforcement and modernisation of Union consumer protection rules (OJ EU L 328 of 2019, No. 328, p. 7, as amended); Act of 23 August 2007 on counteracting unfair market practices ( i.e. Journal of Laws of 2023, item 845).

II.            Ensuring compliance with the Digital Services Act (DSA)

a. Implementing a “report content” mechanism: Every review or user-generated content must have an easily accessible button to report suspected illegality or manipulation of the content.

b. Procedure for justifying decisions: In the event of deletion of an opinion or blocking of a user account, the platform is obliged to send the author a detailed justification indicating a specific violation of the regulations or legal provisions.

c. Internal Complaints Process: Users must be able to appeal moderation decisions for a period of at least 6 months from the date the platform takes action.

d. Designation of a contact point: The entrepreneur must designate an electronic contact point for supervisory authorities and users, enabling efficient communication on matters relating to digital security.

Legal basis: Regulation (EU) 2022/2065 of the European Parliament and of the Council of 19 October 2022 on the single market for digital services and amending Directive 2000/31/EC (Digital Services Act) (OJ EU L 277, 2022, No. 277, p. 1, as amended), in particular Articles 16, 17 and 20.

III.            Reducing the risk of “algorithmic exclusion”

a. Design Patterns (UX) Audit: Eliminate so-called dark patterns, such as asymmetric selector buttons, hard-to-close pop-ups, or mechanisms that make it difficult to unsubscribe. Supervisory algorithms treat such practices as signals of poor interface quality.

b. Data Certification for AI: Ensure structured review data is provided, allowing shopping assistants and crawlers to properly verify the “digital provenance” of the data.

c. Filtering synthetically generated content: It is worth implementing tools that monitor review language for bot-like patterns (unnatural correctness, lack of detail) to avoid indexing false enthusiasm that results in lower trust rankings.

Legal basis: REGULATION (EU) 2022/2065 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 19 October 2022 on the single market for digital services and amending Directive 2000/31/EC (Digital Services Act) (OJ EU L 277, 2022, p. 1, as amended) – Article 25 (prohibition of deceptive interfaces)

IV.            Proper management of data and opinions (CaaS model)

a. Digital Audit Trail: It is recommended to store logs containing transaction metadata related to opinions for a period enabling verification of data reliability (e.g. 12-24 months).

b. Active mediation systems: Instead of deleting negative feedback, use complaint management systems that document the process of resolving customer disputes. Resolving a problem is treated by ranking systems as evidence of high-quality service.

c.       “Know Your Business Customer” principle: When running a marketplace model, it is essential to verify the identity of sellers before allowing them to offer goods, collecting registration numbers and contact details.

Legal basis: REGULATION (EU) 2022/2065 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 19 October 2022 on the single market for digital services and amending Directive 2000/31/EC (Digital Services Act) (OJ EU L of 2022, No. 277, p. 1, as amended) – Article 30; Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) (OJ EU L of 2016, No. 119, p. 1, as amended).

Artykuł Is Your Online Store Ready for the New Era of Control? A Practical Guide to E-Commerce Responsibilities in 2026 pochodzi z serwisu KIELTYKA GLADKOWSKI LEGAL | CROSS BORDER POLISH LAW FIRM RANKED IN THE LEGAL 500 EMEA SINCE 2019.

The dynamic development of artificial intelligence-based technologies is revolutionizing not only the commercial sector but also the area of state oversight of the digital market. The implementation of multi-agent systems by the Office of Competition and Consumer Protection (UOKiK) opens a new era in consumer rights enforcement, enabling the mass and automated identification of unfair market practices. With the Digital Services Act (DSA) and the Omnibus Directive in force, traditional control methods are giving way to algorithmic interface analysis aimed at eliminating so-called dark patterns and price manipulation. However, the use of “digital controllers” raises fundamental questions for legal science and business practice about the limits of automated decision-making processes in public administration. Although AI agents significantly improve the effectiveness of detecting violations, their legal status as a source of evidence remains the subject of heated debate. The main thesis is that while AI can be a powerful auxiliary tool for regulatory bodies, the ultimate responsibility for determining the facts and assessing the legitimate interests of a party must rest with humans, which is the foundation of a fair procedure in a state governed by the rule of law.

Dark Patterns: Legal and Ethical Aspects of Prohibiting Manipulation in Digital Interfaces

A key obligation of internet platform providers in light of modern regulations is to design interfaces in a transparent and ethical manner. The prohibition of manipulation, formulated, among others, in the Digital Services Act (Article 25), directly affects the structure of so-called deceptive interfaces (dark patterns). Websites and applications cannot be designed in a way that limits the recipient’s cognitive autonomy, interferes with their ability to rationally assess the situation, or forces them to make a purchasing decision that they would not have made under other circumstances.

One of the most glaring examples of such violations is the asymmetry in the contract conclusion and termination process, particularly evident in subscription models. This mechanism relies on extreme simplification of the purchase path while simultaneously mounting procedural barriers when attempting to cancel the service. Visual techniques are used here, among other things: payment activation buttons are highlighted with bright colors and a central location, while contract termination options are deliberately hidden at the bottom of the page, written in small font or masked with colors that blend with the background. Furthermore, canceling a subscription on online platforms often requires multiple selections or confirmation of the desire to cancel, despite the consumer’s prior explicit choice. Artificial intelligence algorithms, analyzing the page structure and visual hierarchy of elements, can pinpoint these disparities with mathematical precision, creating a list of violations that serves as hard evidence.

In the context of the Omnibus Directive, the obligation to disclose the lowest price 30 days before the discount has become a market standard, but its implementation is open to abuse. The practice of “empty promotions” involves artificially inflating the base price just before a planned discount or providing a false reference amount. In this area, AI agents demonstrate particular effectiveness, acting as real-time monitoring systems; they can archive the price history of each product, creating an independent database. Comparing this information with the entrepreneur’s declaration visible on the website allows for immediate detection of manipulation of the promotional algorithm.

An equally important area of control is the phenomenon of drip pricing , or hiding the real costs of a transaction until the final stage of the shopping cart. Businesses often employ a “decoy” strategy, presenting an attractive unit price, which, at the time of order finalization, is increased by mandatory, previously undisclosed costs, such as service fees, packaging costs, or payment processing fees. Pursuant to Article 12 of the Consumer Rights Act, businesses are obligated to clearly and understandably inform consumers about, among other things, the total price for the proposed service. Automated control systems are capable of conducting a full simulation of the purchasing process, from product selection to the payment gateway. Any discrepancy between the price presented in the product list and the amount required to complete the contract is reported by AI as an attempt to circumvent disclosure obligations and a direct violation of the collective interests of consumers.

According to Article 5 of the Act on Combating Unfair Market Practices, the key criterion for assessing a trader’s behavior is the impact of their actions on the recipient’s decision-making process. A market practice is considered misleading if “this action in any way causes or is likely to cause the average consumer to make a transactional decision that they would not otherwise have made”. The legislator specifies that both “spreading false information” and “spreading true information in a manner that is likely to be misleading” can constitute an infringement. In the digital environment, these manipulations most often focus on the “existence of a product, its type, or availability.” A common method of exerting unjustified pressure on consumers is the use of social proof mechanisms and an artificial sense of scarcity. This manifests itself in messages such as: “this product is now being viewed by x people,” “x items have already been purchased today,” or displaying timers indicating that “only 30 minutes left until the end of the promotion.” Particularly problematic from the perspective of trade ethics is the use of so-called false advertising. Timers – clocks counting down to the finale of a supposedly unique price opportunity. In reality, these are fake mechanisms, as after the specified deadline, the offer remains active and the product price remains unchanged or becomes even more favorable. This type of activity, a classic example of dark patterns, is designed to induce fear of missing out (FOMO) in customers and induce them to rush into a transaction. Using AI agents allows regulators to serially monitor such counters and prove their cyclical recurrence, providing direct evidence of deceptive practices.

The algorithm as a controller

With millions of transactions taking place across the country in just a few minutes or hours, standard order verification procedures prove insufficient to effectively fulfill the statutory responsibilities of supervisory authorities. Technological advancements in the form of AI algorithms come to the rescue. These algorithms can automatically monitor numerous commercial transactions simultaneously, generating preliminary opinions that are ultimately subject to human review. Such systems not only save significant processing time but, above all, enable oversight of a much broader range of businesses and their online platforms. The AI multi-agents used in this process are virtual “consumer robots” capable of mass-auditing e-commerce websites, simulating the natural behavior of online users to detect irregularities that a human controller would be unable to detect on such a large scale.

To conduct reliable and effective inspections, Polish law already offers supervisory authorities a toolkit in the form of the “mystery shopper” institution. Traditionally, this involves a person unrelated to the inspected company or the inspecting authority making a purchase and then completing a survey regarding specific activities they observe during standard shopping. The implementation of AI technology by the Office of Competition and Consumer Protection (UOKiK) aims to entrust AI multi-agents with the role of such digital “mystery shoppers.” Their task is to interact with the website interface, add a product to the cart, and complete the entire purchasing process without disclosing that this activity is being performed by an algorithm or that it is part of an official inspection procedure. This approach allows for direct verification of whether the entrepreneur is not using prohibited manipulative practices, known as dark patterns. However, it should be emphasized that the activity of AI multi-agents is strictly regulated by legal procedures and cannot be arbitrary. The algorithm operates under the strict supervision of the President of the Office of Competition and Consumer Protection, who, pursuant to Article 105ia of the Act on Competition and Consumer Protection, must always obtain prior consent from the Court of Competition and Consumer Protection. This mechanism serves as a key safeguard against abuse of power. Furthermore, after completing the inspection, the office is obligated to immediately provide the entrepreneur with an official ID and authorization for the inspection. In the age of digital administration, this obligation can be fulfilled electronically immediately after the AI multi-agents withdraw from the sales platform.

The key legal framework for the operation of algorithms commissioned by the regulator is provided by the EU AI Act. According to its provisions, AI systems used by public authorities for control and supervisory purposes should be considered high-risk AI systems. This entails a strict requirement to design them with appropriate transparency, which allows both the controlling and the controlled entities to properly interpret the system’s results and use them fairly. In practice, this means that algorithms must be built in an “explainable” model. A business subject to allegations based on an algorithmic audit has the statutory right to request full insight into the operation of AI tools. This transparency is essential for the controlled entity to understand the basis and criteria on which the authority deemed its online platform unfair or infringing on the collective interests of consumers (Article 24). This balance between the effectiveness of digital supervision and the right to defense is the foundation of a modern rule of law in the age of algorithms.

The opinion of AI multi-agents as evidence in the case

After completing the inspection activities on the entrepreneur’s online platform, the AI algorithm’s role evolves towards an analytical function, consisting of preparing an opinion indicating detected violations. In the context of potential proceedings against an entity employing unfair market practices, the admissibility of using such an analysis as valid evidence becomes a key issue. Pursuant to Article 7 of the Code of Administrative Procedure (hereinafter referred to as the Code of Administrative Procedure), which establishes the principle of objective truth, a public administration body is obligated to take all steps necessary to thoroughly clarify the factual circumstances. This obligation is consistent with Article 75 § 1 of the Code of Administrative Procedure, which introduces an open catalog of evidence, allowing as evidence anything that may contribute to the clarification of the case, provided it is not contrary to the law.

Under these regulations, the results of AI multi-agent work – taking the form of reports, opinions, or analyses generated after conducting an audit with court approval – fully fall within the statutory definition of evidence. However, it should be clearly stated that an AI opinion cannot be equated with an expert opinion within the meaning of Article 84 of the Code of Administrative Procedure. This stems from the fact that an algorithm does not possess the status of a natural person equipped with specialized knowledge, which is a statutory requirement for appointing an expert. Instead, documentation generated by an AI agent should be classified as a private document or so-called “unnamed evidence.”

Practical justification for this position can be found in the case law concerning digital evidence. The judgment of the Court of Appeal in Szczecin of September 19, 2016, I ACa 364/15, LEX no. 2147337 aptly describes this issue, pointing out that evidence in a case may include official and private documents, but also means other than those listed in Articles 305-308 of the Code of Civil Procedure. Electronic evidence, currently increasingly used in civil proceedings, is not explicitly listed in the catalog of means of evidence. However, the Code of Civil Procedure does not contain a closed list of evidence sources; anything relevant to the case may constitute evidence. Although the above ruling was issued in the context of civil procedure, due to the identical approach to the openness of the evidence system, it remains fully applicable to administrative proceedings conducted by the President of the Office of Competition and Consumer Protection.

The key element of algorithmic evidence remains the human factor, which serves as a primary safeguard over the autonomous operation of technology. It’s important to note that AI multi-agents, despite their high sophistication, operate based on statistical probability models, which carries the risk of misinterpreting dynamic website elements. For example, the system may incorrectly classify a standard technical error as intentional dark web activity. patterns or misinterpret the interface’s intentions in a specific cultural or linguistic context. Therefore, opinions generated by AI agents cannot constitute a standalone and final basis for a decision, but should be subjected to thorough, critical review by an official. Only such a comparison of the “raw” algorithmic result with human knowledge and experience allows for avoiding errors that could lead to unjustified penalties. This approach is directly supported by Article 80 of the Code of Administrative Procedure, according to which a public administration body assesses whether a given circumstance has been proven based on the entirety of the evidence. In this process, the “AI opinion” is only one of many components that must be weighed against other evidence and evaluated through the prism of principles of logic and life experience, ultimately guaranteeing the implementation of the principle of objective truth and protecting the entrepreneur from the automaticity of decisions made by the algorithm.

Summary

Multi-agent system implemented by the Office of Competition and Consumer Protection for automatic control of the e-commerce sector poses a significant challenge for entrepreneurs, forcing strict compliance with regulations regarding dark patterns, price transparency (Omnibus Directive, Art. 6a) and information obligations (Consumer Rights Act, Art. 12). These tools are used to mass detect manipulative practices such as drip pricing, fake timers or making it difficult to unsubscribe. Although AI agents perform a function similar to “mystery shoppers,” their activity must meet the rigors of Article 105ia of the Act on Competition and Consumer Protection, including the requirement to obtain court consent for a controlled purchase. What is crucial from a procedural perspective is that the findings made by the algorithm do not have the status of an expert opinion within the meaning of Article 84 of the Code of Administrative Procedure (lack of the status of a natural person with specialist knowledge), but constitute only a private document or “other evidence” subject to the authority’s free assessment (Article 80 of the Code of Administrative Procedure).

Consequently, the official is required to subject AI reports to thorough human review to eliminate the risk of misclassification resulting from so-called “AI hallucinations” or technical errors in the interpretation of the website’s code. The entrepreneur has full rights of defense based on the principle of active participation of the party (Article 10 of the Code of Administrative Procedure) and the principle of objective truth (Article 7 of the Code of Administrative Procedure), which means the right to question the bot’s logic and to access the instructions and parameters of the AI system, in accordance with the “explainability” requirement enshrined in the AI Act (Article 13). Any decision based solely on the automated generation of conclusions, without providing the party with an opportunity to comment on the evidence (Article 81 of the Code of Administrative Procedure), constitutes a gross violation of administrative procedure and may constitute an effective basis for challenging the authority’s decision.

Sources:

Regulation 2022/2065 on the single market for digital services and amending Directive 2000/31/EC (Digital Services Act) (OJ EU L 277, 2022, No. 277, p. 1, as amended).

Directive (EU) 2019/2161 of the European Parliament and of the Council of 27 November 2019 amending Council Directive 93/13/EEC and Directives 98/6/EC, 2005/29/EC and 2011/83/EU of the European Parliament and of the Council as regards the better enforcement and modernisation of Union consumer protection rules (OJ L 328, 2019, p. 7, as amended).

Act of 30 May 2014 on consumer rights (consolidated text: Journal of Laws of 2024, item 1796, as amended).

Act of 23 August 2007 on counteracting unfair market practices (consolidated text: Journal of Laws of 2023, item 845).

Act of 16 February 2007 on competition and consumer protection (consolidated text: Journal of Laws of 2025, item 1714).

Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence and amending Regulations (EC) No 300/2008, (EU) No 167/2013, (EU) No 168/2013, (EU) 2018/858, (EU) 2018/1139 and (EU) 2019/2144 and Directives 2014/90/EU, (EU) 2016/797 and (EU) 2020/1828 (Artificial Intelligence Act) Text with EEA relevance (OJ L 1689, 2024).

Act of 14 June 1960, the Code of Administrative Procedure (consolidated text: Journal of Laws of 2025, item 1691).

Judgment of the Court of Appeal in Szczecin of 19 September 2016, I ACa 364/15, LEX no. 2147337.

Artykuł Multi-agent system in the service of the Polish Office of Competition and Consumer Protection – a new era of e-commerce control and the limits pochodzi z serwisu KIELTYKA GLADKOWSKI LEGAL | CROSS BORDER POLISH LAW FIRM RANKED IN THE LEGAL 500 EMEA SINCE 2019.

The draft act amending the Act on the Provision of Electronic Services and certain other acts represents a significant step in adapting Polish legal regulations to the rapidly evolving digital landscape. The growing number of entities operating online, new methods of electronic communication, and the vast scale of data processing mean that the current regulations are increasingly imprecise or inadequate for the current times. In response to these challenges, the legislator proposed changes aimed at both increasing the protection of network users and clarifying the obligations of entities providing electronic services.

The Act of 18 July 2002 on the Provision of Electronic Services has been the fundamental act regulating electronic services in Poland since the beginning of the digital era. Its provisions included, among other things, the obligations of service providers related to the provision of electronic services, the principles for excluding service providers’ liability in connection with the service provided, and the principles for protecting the personal data of individuals using electronic services.

The 2025 amendment aimed to adapt national law to the requirements of Regulation (EU) 2022/2065 (Digital Services Act – DSA). This regulation has been directly applicable in the European Union since February 17, 2024, but member states – including Poland – were required to introduce provisions into their national legal systems to enable its practical application (e.g., by designating responsible national authorities and enforcement procedures). The implementation of EU regulations also aimed to enhance the protection of service users and their fundamental rights, combat illegal content, and enable state authorities to effectively enforce the law and introduce mechanisms for legal interference online where unnecessary – particularly through administrative decisions by state authorities rather than solely voluntary actions by platforms.

The Act amending the Act changes the so-called reference in the Act’s title, adding that the Act not only implements Directive 2000/31/EC, the Electronic Commerce Directive, but also serves to apply EU Regulation 20222/2065 (Digital Services Act). Before the amendment, the Act referred exclusively to modern content liability rules, which are regulated in the EU by the Digital Services Act (DSA). However, the Act now formally recognizes that it serves to apply the DSA – which is significant because the DSA is a directly applicable regulation but requires national organizational regulations.

Orders to take action against illegal content

A significant element of the amendment is the addition of a new Article 11a et seq. to the Act – within a completely new Chapter 2a titled: “Orders to take action against illegal content, orders to remove restrictions imposed by a hosting service provider, and orders to provide information”. This procedural solution creates a national system of “administrative orders” against service providers and platforms, representing a level and scope of regulation that was absent from the 2002 Act and implicitly provided for in the DSA. Chapter 2a establishes a detailed administrative procedure enabling state authorities to issue binding decisions in cases concerning illegal content and to verify moderation efforts undertaken by platforms. In particular, it introduces:

• Catalogue and definitions of illegal content – the Act refers to specific provisions of criminal law and other specific acts (including the Penal Code, the Act on Copyright and Related Rights, Industrial Property Law and health protection regulations), precisely defining the types of content the dissemination of which may constitute the basis for issuing an administrative order;
• A formalized procedure for submitting applications for the issuance of an order – the application may be submitted by, among others, the Prosecutor, the Police, the National Revenue Administration authorities, the service recipient, as well as so-called trusted whistleblowers in the case of hosting services. This procedure has been linked to the ” notice and action ” mechanism resulting from the DSA, which strengthens its guarantee nature;
• The scope of possible decisions of the authority – the administrative authority may issue an order to prevent access to illegal content, including content that may constitute a prohibited act, as well as an order to remove restrictions imposed by the hosting service provider on the service recipient, if the collected evidence does not confirm the illegal nature of the content;
• Quick and formalized deadlines for considering the case – the legislator has provided for short deadlines for issuing decisions (from 2 to 21 days), and in certain cases the possibility of making the decision immediately enforceable, which significantly increases the effectiveness of the authorities’ interventions;
• Judicial review of decisions – although decisions issued in this mode are not subject to a classic administrative appeal, the parties have the right to lodge an objection heard by a common court, which serves as a guarantee and is an important element of protecting the rights of service recipients and providers of indirect services.

Such regulations were absent from the 2002 Act, which was based primarily on general principles of service user liability and voluntary content moderation mechanisms employed by platforms. The amendment created a formal prescriptive procedure, which in practice grants state authorities the actual authority to decide – through administrative decisions – to block specific content or restore access to it, significantly changing the current model for regulating services provided electronically.

Competent authorities

The newly introduced Article 3a of the Act, which is titled “Competent authorities, the coordinator for digital services and the National Council for Digital Services”, lists the competent state authorities and their competences:

• The President of the Office of Competition and Consumer Protection is listed as the competent authority in matters
  • breaches of obligations by internet platform providers under Chapter III of Sanctions 4 of the DSA
  • Other breaches of the DSA Regulation resulting in harm to consumer interests.
• The Chairman of the National Broadcasting Council is the competent body in matters relating to:
  • Video sharing platform
  • Excluding matters falling within the competence of the President of the Office of Competition and Consumer Protection under the DSA
• The President of the Office of Electronic Communications is a subsidiary competent authority – it deals with all other matters within the scope of the DSA.
• Competence disputes are resolved by the Prime Minister

In matters not regulated, the Code of Administrative Procedure applies. With respect to certain decisions, the possibility of challenging a decision under the extraordinary procedures of the Code of Administrative Procedure (renewal, invalidation, amendment, or repeal) is excluded.

The Polish President of the Office of Electronic Communications (UKE) has been appointed as the Digital Services Coordinator, serving as the central point of contact for national and EU relations. His responsibilities include representing Poland in the European Digital Services Council (EDSC), whose meetings are also attended by the President of the Office of Competition and Consumer Protection (UOKiK) or the Chairman of the National Broadcasting Council (KRRiT), depending on when matters within their jurisdiction are discussed. He collaborates with the President of the UOKiK and the Chairman of the National Broadcasting Council (KRRiT) in fulfilling obligations arising from the Act and the Digital Services Directive (DSA), and ensures the consistency of the national application of DSA provisions. The Coordinator is also responsible for preparing a single, consolidated report covering the activities of all relevant authorities, with these authorities required to submit annual reports by 31 March each year.

National Council for Digital Services

The National Digital Services Council operates under the Digital Services Coordinator, serving as an advisory and consultative body on matters related to the functioning of the digital services market. The Council’s mission is to support the creation of a safe, predictable, and trustworthy digital environment, in particular by providing opinions on the implementation of intermediary service providers’ obligations under the Digital Services Act (DSA), the functioning of trusted whistleblowers, and alternative dispute resolution systems, as well as by formulating proposals regarding data access for verified researchers. The Council consists of a Chairperson, a Deputy Chairperson, and ten members appointed by the President of the Office of Electronic Communications (UKE) from among representatives of the scientific community, social and economic organizations, and entities operating in the digital services market. The Council’s term of office is four years, and its administrative support and funding are provided by the office serving the UKE.

The system is complemented by the possibility of financial support for entities playing a significant role in implementing the mechanisms provided for in the DSA. The Act provides for the awarding of targeted subsidies to both out-of-court dispute resolution bodies and trusted entities reporting illegal content, through open and non-discriminatory recruitment. This solution aims to strengthen the practical effectiveness of the user protection system and the enforcement of obligations imposed on digital service providers.

The amendment also introduces chapters 4a-4f which regulate the certification of out-of-court dispute resolution bodies, the status of a trusted whistleblower, the status of a verified researcher, the liability of intermediary service providers, civil liability and fines.

Certification of out-of-court dispute resolution bodies

Certification of alternative dispute resolution bodies is provided by the Digital Services Coordinator for a five-year period, in accordance with the DSA rules. The entity seeking certification submits an electronic application, signed with a qualified, trusted, or personal signature, containing, among other things, its name, address, the scope of its expertise, and the languages in which it can resolve disputes. The application must be accompanied by documents confirming compliance with certification requirements.

Before issuing a certificate, the coordinator consults with public administration authorities to confirm competence in matters related to illegal content or the application and enforcement of terms of service. The application is reviewed within two months; if any deficiencies are found, the coordinator requests supplementation within seven days.

After a positive assessment, the coordinator issues a certificate that includes the name of the authority, validity period, certificate number, and the coordinator’s signature. During the certificate’s validity period, the authority must meet the certification requirements. The certificate may be extended for another five years upon request by the authority, submitted no later than 30 days before its expiry, along with a declaration of compliance.

The Coordinator maintains a list of certified bodies, published in the Public Information Bulletin. A certificate is refused or revoked by decision if the body fails to meet the requirements. The decision is immediately enforceable and subject to appeal to the administrative court.

The coordinator may conduct verification activities to ensure compliance with certification requirements, including access to the body’s premises, reviewing documents, and requesting clarification. A report is prepared for the activities, signed by the coordinator and the body. Certified bodies submit annual reports to the coordinator by March 31 of the following year, with the possibility of requesting supplementation.

Trusted whistleblower status

Trusted Signaling entity status is granted by the Digital Services Coordinator in accordance with the DSA rules. Applicants submit an electronic application with the required information: name, address, email address, areas of expertise, and documents confirming compliance with the requirements of the EU regulation.

Before granting the status, the coordinator will seek the opinion of the President of the Personal Data Protection Office regarding the data protection measures applied by the entities, as well as the opinion of other public administration bodies regarding their expertise in illegal content or the application of the terms and conditions of use of online platforms. These bodies have 30 days to submit their opinion; failure to respond within this period will be deemed to have met the requirement.

The coordinator reviews the application within two months and requests supplementation if any deficiencies are found. Upon positive evaluation, the coordinator issues a certificate that includes the entity’s name, number and date of issue, the scope of expertise, and the coordinator’s signature. The status and information about trusted entities are published in the Public Information Bulletin.

The coordinator may refuse to grant status if the entity does not meet the requirements, and the decision is final. To verify the conditions, they may conduct inspections, suspending the status for the duration of the inspection. The status may be revoked if the entity no longer meets the requirements, and the decision is immediately enforceable. A complaint may be filed against a decision refusing or revoking status to the administrative court.

Trusted flaggers are obliged to submit annual reports to the coordinator on reports made in accordance with Article 16 of the EU Regulation, by 31 March of the following year.

Verified Researcher Status

Verified researcher status, which allows access to data for research purposes (Article 40, paragraph 4 of the DSA), is granted by the Digital Services Coordinator, i.e., the President of the Office of Electronic Communications. An entity or individual applying for this status submits an application containing: name, first name and last name, registered office or place of residence address, email address, information confirming compliance with the requirements of Article 40, paragraph 8 of the Regulation, and a description of appropriate technical and organizational security measures, along with supporting documents.

Before issuing a decision, the coordinator will seek the opinion of the President of the Personal Data Protection Office (UODO) and other public administration bodies, assessing whether the planned research contributes to the detection and understanding of systemic risk and whether the scope and conditions of data access are adequate and proportionate. The bodies have 30 days to submit their opinion; failure to respond will be deemed to have met the requirement.

The coordinator reviews the application within two months, requesting that any missing information be provided within seven days; otherwise, the application will not be considered. Upon positive evaluation, the coordinator issues a certificate confirming the status of a verified researcher. Applications are submitted to providers of very large online platforms or search engines (VLOP and VLOSE) via a single point of contact.

At the request of a platform or search engine, the coordinator may approve or deny an application’s amendment, and this decision is immediately enforceable. The coordinator refuses to grant status if the applicant does not meet the requirements, and the decision is final. To verify the conditions, the coordinator may conduct verification activities with the applicant and the entity to which the status was granted.

Access to data may be revoked if requirements are not met or regulations are violated, and the decision is immediately enforceable. The coordinator then informs the platform or search engine provider and withdraws the access request. An appeal against the coordinator’s decision may be filed with the administrative court.

Liability of indirect service providers

A service recipient or other authorized entity may submit a complaint against an indirect service provider to the Digital Services Coordinator. The complaint should include the provider’s details, the service, a description of the violation, the legal basis, and evidence (documents, screenshots, recordings). The Coordinator will inform the complainant about the complaint’s handling and possible transfer to an authority in another EU country, and, if necessary, request translation of documents.

Proceedings before the competent authority may be explanatory (to determine whether there are grounds for further proceedings) or concerning a breach of duty. The authority may initiate them ex officio, and the explanatory proceedings are concluded within 4-5 months.

During the proceedings, the authority may conduct an inspection of the inspected entity and other individuals with information related to the violation. The inspection includes access to the premises, review of documents and systems, requesting clarification, making copies, and securing materials. The inspector may be supported by the Police and NASK-PIB.

The inspected entity is obligated to provide information, ensure the conditions for the inspection, allow access to documents, systems, and premises, and confirm the conformity of copies with the originals. Evidence may be secured at the entity’s premises or at an office. If necessary, the competent authority may issue a decision to seize documents or items for the period necessary for the inspection (max. 7 days).

Once a report of seized items is prepared, they can be collected or transferred to a trusted person, with the obligation to present them to the authority upon request. Inspection and seizure reports document, among other things, the case details, the list of items, the course of action, and the signatures of the inspector and the inspected party. The inspected party may submit objections to the report, which the authority analyzes and, if necessary, supplements or amends the report in the form of an annex. Inspections can be conducted remotely, and all information obtained during the inspection, including trade secrets, is protected. Legally protected materials, such as correspondence with a lawyer, remain at the inspection site, and in case of doubt, they are forwarded to the Competition and Consumer Protection Court.

The competent authority may conduct inspections of providers of large platforms and search engines, and in the event of violations of the DSA Regulation, it issues decisions ordering cessation of the infringement, publication of the decision, warnings, or restrictions on access to services, while maintaining the proportionality of the measures. If the provider ceases the infringement, the decision establishes this fact, and the burden of proof rests with the provider. The authority may impose remedial obligations, monitor their implementation, and issue interim measures if there is a risk of serious harm.

Suppliers are obligated to provide information and documents upon request by the authority, while respecting the right of individuals to refuse in certain situations. The authority’s decision may be appealed to the Court of Competition and Consumer Protection, which the authority forwards along with the case file, retaining the option of prior revocation or amendment of the decision. Regarding evidence and service, the provisions of the Code of Civil Procedure and regulations on electronic service apply accordingly.

Civil liability and court proceedings

In matters concerning claims by service recipients for breach of obligations arising from the DSA, the provisions of the Civil Code and the Code of Civil Procedure apply to matters not regulated by the regulation. The competent court for resolving claims by service recipients is the district court. The court notifies the competent authority of the filing of a lawsuit and the final judgment, and the authority informs the court of pending or concluded proceedings concerning the violation. The court suspends the proceedings if the authority upholding the claim discontinues the legal proceedings in this regard. The authority’s findings regarding the breach of obligations are binding on the court when assessing liability for damages. The competent authority or trusted whistleblower, with the service recipient’s consent, may bring legal proceedings on their behalf and participate in the proceedings at any stage, with the provisions of the Code of Civil Procedure applying accordingly. The authority may also submit opinions to the court on matters of public interest related to the claim arising from the violation of the regulation.

Fines

Providers of intermediary services, internet platform hosting, and internet search engines who fail to comply with the obligations set out in the DSA are subject to administrative fines imposed by the competent authority. Fines can amount to up to 6% of the global turnover of the supplier in the year preceding the imposition of the fine. In the event of improper provision of information to the authority or obstruction of an inspection, the fine can amount to up to 1% of annual income or turnover, and an additional periodic penalty of up to 5% of average daily turnover can be imposed for each day of delay or obstruction in fulfilling the obligations. Fines may be imposed even after the infringement has ceased, if the duration, scope, or effects of the infringement warrant.

The turnover amount required to establish the fine is calculated based on the profit and loss account, financial statement, or other revenue documents, also taking into account the turnover of the merged entities or the average turnover in recent years. In the absence of revenue, the maximum fine is the equivalent of EUR 6,000. The euro is converted to PLN at the NBP exchange rate on the last day of the year preceding the penalty.

When determining the amount of fines, the authority is guided by the public interest and the principles of effectiveness, proportionality, and deterrence, taking into account the nature and gravity of the violation, the type of business activity, and the economic capacity of the entity. Proceeds from fines constitute state budget revenue. Fines are enforced pursuant to the provisions on administrative enforcement proceedings, and payment is due within 14 days of the decision becoming final. In unregulated matters, the provisions of the Tax Ordinance apply accordingly. Fine decisions may be appealed to the Court of Competition and Consumer Protection.

In January 2026 the implementation of the EU Digital Services Act (DSA) in Poland was blocked by a presidential veto of a key amendment to the act that was to implement provisions on blocking illegal content by the administration (the President of the Office of Electronic Communications), raising concerns about leaving internet users without full protection and shifting responsibility to the European Commission; the president justified the veto with concerns about “administrative censorship” and “overregulation.”

The bill, as submitted for the President’s signature, included amendments resulting from a broad public discussion on the need to increase control over digital platforms.

The adopted amendments, among other things, increased judicial oversight of the procedure for blocking illegal content by limiting the automatic immediate enforceability of decisions against which an appeal has been filed.

They also expanded the list of entities that may apply for an injunction to take action against illegal content to include copyright and related rights holders (even if they are not users of the platform in question).

They also clarified that the injunction to unblock content applies to situations in which the provider has made an erroneous decision (often using automated moderation tools), not to situations where standard terms of service have been applied.

Artykuł Poland blocks the national implementation of the EU Act on the provision of electronic services pochodzi z serwisu KIELTYKA GLADKOWSKI LEGAL | CROSS BORDER POLISH LAW FIRM RANKED IN THE LEGAL 500 EMEA SINCE 2019.