THE EUROPEAN MODEL OF DI= GITAL PLATFORM REGULATION:=

DSA AND AI = ACT AS A REFERENCE FRAMEWORK FOR BRAZIL=

O modelo europeu de regulação de plata= formas digitais:

DSA e AI Ac= t como referência par= a o Brasil

Pawel Kowalski*

Resumo: This article analyzes the European model of digital platform regulat= ion, embodied in Regulation (EU) 2022/2065 (Digital Services Act) and Regulation (EU) 2024/1689 (AI Act), as a reference framework for the Brazilian legal system following the Supreme Federal Court decision that declared the parti= al unconstitutionality of Article 19 of the Marco Civil da Internet on June 26, 2025. The research examines the evolution of digital intermediary liability= in European law, from the E-Commerce Directive of 2000 to the current tiered d= ue diligence system of the DSA, as well as the algorithmic transparency obligations imposed by the AI Act. The study identifies elements of the European model potentially applicable to the Brazilian context, considering constitutional specificities and the categories of unlawful content establi= shed by the STF. It concludes that the European experience offers relevant insig= hts for future Brazilian legislation, particularly regarding systemic risk management, content moderation, and artificial intelligence transparency.

Keywords: digital platforms; Digital Services Act; AI Act; Marco Civil da Internet; intermediary liability; content moderation; artificial intelligen= ce.

Resumo: O presente artigo analisa o modelo europeu de regulaçã= ;o de plataformas digitais, consubstanciado no Regulamento (UE) 2022/2065 (Digital Services Act) e no Regulamento= (UE) 2024/1689 (AI Act) como marco de referência para o ordenamento jurídico brasileiro= após a decisão do Supremo Tribunal Federal que declarou a inconstitucionalidade parcial do artigo 19 do Marco Civil da Internet em 26 de junho de 2025. A pes= quisa examina a evolução da respon= sabilidade dos intermediários = digitais no direito europeu,= desde a Diretiva de Comércio Eletrônico de 2000 até o atual sistema de devida diligência escalonada do DSA, bem como as obrigações de tran= sparência algorítmica imposta= s pelo AI Act. O estudo identifica elementos do <= span class=3DSpellE>modelo europeu potencialmente aplicáveis= ao contexto brasileiro, considerando as especifici= dades constitucionais e as categ= orias de ilícitos estabel= ecidas pelo STF. Conclui-s= e que a experiência europeia oferece subsídios relevantes para a futura legislaç&a= tilde;o brasileira, especialmente<= /span> no tocante à gest&a= tilde;o de riscos sistêmicos= , moderação de conteúdo e transparência dos = sistemas de inteligência artificial.

Palavras-chave: plataformas digitais; Digital Services Act; AI Act;= Marco Civil da Internet; responsabilidade de intermediários; modera&cc= edil;ão de conteúdo; inteli= gência artificial.

INTRODUCTION

T= he regulation of digital platforms constitutes one of the major legal challeng= es of the twenty-first century. The economic and social power concentrated in large technology companies, combined with the sp= eed of content dissemination on the internet, requires regulatory responses that balance the protection of fundamental rights with the preservation of freed= om of expression and technological innovation.

In Brazil, this regulatory tension manife= sted paradigmatically in the judgment of Extraordinary Appeals RE 1.037.396 (The= me 987) and RE 1.057.258 (Theme 533) by the Supreme Federal Court, concluded on June 26, 2025. By a majority of eight votes to three, the Court declared the partial unconstitutionality of Article 19 of Law No. 12,965/2014 (Marco Civ= il da Internet), which established that internet application providers could o= nly be held civilly liable after failing to comply with a specific judicial ord= er determining the removal of content.

The STF decision represents a significant= shift in the Brazilian model of intermediary liability, bringing it closer to regulatory trends observed in other jurisdictions, especially in the Europe= an Union. The Court recognized that the general rule of Article 19 does not provide sufficient protection for constitutional legal interests of high relevance, notably the protection of fundamental rights and democracy.=

Meanwhile, the European Union has consoli= dated over the past two decades a sophisticated normative framework for the governance of the digital environment. Regulation (EU) 2022/2065, known as = the Digital Services Act (DSA), and Regulation (EU) 2024/1689, called the AI Ac= t, represent the central pillars of this regulatory architecture, establishing differentiated obligations according to the size and nature of digital services, as well as specific transparency requirements for artificial intelligence systems.

The purpose of this article is to analyze= the European model of digital platform regulation as a reference framework for = the development of Brazilian law in this matter. The research is justified by t= he need to support the legislative debate that will follow the STF decision, as well as by the relevance of understanding comparative experiences in a rapi= dly evolving legal field.

Methodologically, the study adopts the perspective of functional comparative law, seeking to identify not only normative similarities and differences, but above all the common problems f= aced and the solutions developed in each legal order. The analysis focuses on the following axes: (i) the historical evolution of intermediary liability in European law; (ii) the tiered due diligence syste= m of the DSA; (iii) the algorithmic transparency obligations of the AI Act; and = (iv) the possibilities for transposition to the Brazilian context.

<= span lang=3DEN-US style=3D'mso-fareast-font-family:"Times New Roman";mso-bidi-fo= nt-family: "Times New Roman";mso-ansi-language:EN-US'>= 1 = THE EVOLUTION OF INTERMEDIARY LIABILITY IN THE EUROPEAN UNION

= 1.1 From the E-Commerce Directiv= e to the Digital Services Act

T= he regulation of digital intermediaries in the European Union has its origins = in Directive 2000/31/EC, known as the E-Commerce Directive (European Union, 20= 00). Adopted in a context of optimism regarding the democratizing potential of t= he internet, the Directive established a regime of conditional liability exemp= tion that remained in force for more than two decades.

The model of the 2000 Directive was based= on the distinction between three categories of intermediary services: mere conduit, caching, and hosting. For each category, specific conditions for liability exemption were established, the most rele= vant for digital platforms being the rule of Article 14, which exempted hosting service providers from liability for stored content, provided they did not = have actual knowledge of illegal activity or information (European Union, 2000, = Art. 14).

The concept of "actual knowledge&quo= t; became the central axis of the European liability regime, generating intense doctrinal and jurisprudential controversy over its scope and application. As Keller (2018, p. 294) observes, "the law of intermediary liability lim= its OSPs' legal responsibility for user activities and effectively protects individual Internet users' rights to seek and impart information."

The absence of harmonization of notice-and-takedown procedures across Member States resulted = in regulatory fragmentation, creating legal uncertainty for both service provi= ders and affected rights holders. The Court of Justice of the European Union addressed this issue in landmark cases such as Google France (C-236/08), establishing that the hosting exemption applies when the service provider's role is "neutral, in the sense that its conduct is merely technical, automatic and passive" (CJEU, 2010, para. 114).

Throughout the 2000s and 2010s, the transformation of the digital ecosystem revealed the limitations of the original model. Platforms ceased to be mere passive hosts of third-party content to assume an active role in organizing, curating, and recommending information through sophisticated algorithms that shape user experience and influence public debate. Events such as the disinformation crisis during th= e covid-19 pandemic, interference in electoral processes, and the proliferation of hate speech catalyzed the demand for a profound reform of the regulatory framewo= rk.

In December 2020, the European Commission presented the Digital Services Act proposal, which would be adopted as Regulation (EU) 2022/2065 on October 19, 2022 (European Union, 2022). The D= SA entered into force on November 16, 2022, with staggered application: obligations for Very Large Online Platforms (VLOPs) and Very Large Online Search Engines (VLOSEs) — those with more = than 45 million monthly active users in the European Union — became applic= able from August 25, 2023; the remaining provisions applied from February 17, 20= 24.

<= span lang=3DEN-US style=3D'mso-fareast-font-family:"Times New Roman";mso-bidi-fo= nt-family: "Times New Roman";mso-ansi-language:EN-US'>= 1.2 Structural principles of the= DSA

The DSA is founded on three structural principles that represent a significant evolution from the previous model. The first is the principle of due dilige= nce, which replaces the reactive logic of conditional exemption with a proactive risk management approach. As stated in Recital 79 of the DSA, "given t= he particular risks posed by very large online platforms and very large online search engines for the dissemination of illegal conte= nt and for societal harms, specific rules should be established for such provi= ders" (European Union, 2022, Recital 79).

The second principle is that of proportionality or tiered approach, according to which regulatory obligations are calibrated according to the size, nature, and social impact of services. This principl= e recognizes that it would not be reasonable to impose on small businesses the same requirements applicable to large global platforms, while ensuring that grea= ter market power corresponds to greater responsibility. Article 33 of the DSA establishes that platforms reaching 45 million monthly active users shall be designated as VLOPs or VLOSEs and subject to enhanced obligations (European Union, 2022, Art. 33).

The third principle is transparency, which permeat= es the entire regulation and manifests itself in multiple dimensions: transpar= ency of terms of service, content moderation systems, recommendation algorithms,= and individual decisions to remove or restrict content. Article 27 requires onl= ine platforms to provide recipients with information about parameters used in recommender systems, including "the main parameters used in their recommender systems, as well as any options for the recipients of the servi= ce to modify or influence those main parameters" (European Union, 2022, A= rt. 27).

<= span lang=3DEN-US style=3D'mso-bidi-font-size:12.0pt;line-height:150%;mso-fareas= t-font-family: "Times New Roman";mso-bidi-font-family:"Times New Roman";mso-ansi-language: EN-US'>1.3 Maintenance of liability exemptions

It is important to note that the DSA does not repeal the liability exemptions established by the E-Commerce Directive. Articles 4 to= 6 of the DSA reproduce, with technical improvements, the conditions for exemp= ting service providers from liability for mere conduit, caching, and hosting services.

Article 6 of the DSA, in particular, maintains the rule that hosting service providers are not liable for information stored at the request of a recipient of the service, provided that: (a) they do not h= ave actual knowledge of illegal activity or content; or (b) upon obtaining such knowledge, they act expeditiously to remove or disable access to the illegal content (European Union, 2022, Art. 6).

The innovation of the DSA lies not in altering the= se basic exemptions, but in creating an extensive set of due diligence obligat= ions that condition the maintenance of the exemption and establish standards of conduct for all digital intermediaries. In this sense, the DSA can be understood as an additional regulatory layer that is superimposed on the exemption regime without replacing it.

&= nbsp;

2.1 Architecture of tiered obligations<= /span>

The DSA establishes an architecture of obligations organized in four successive layers, each applicable to a category of service providers. The first layer, the most basic, applies to all intermediary services. The second layer adds specific obligations for hosting services. The third layer adds requirements for online platforms. The fourth and most demanding layer imposes enhanced obligations on very large online platforms and search engines.

This tiered structure ensures that small and medium-sized enterprises are not burdened with disproportionate requirement= s, while subjecting actors with greater systemic impact to more rigorous obligations. As Buiten (2021) observes, the DSA "proposes new, asymmet= ric obligations, while maintaining the liability exemption for hosting providers," representing a shift from liability to regulation.

2.2 General obligations for all intermediaries

The obligations applicable to all intermediary services (Articles 11 to 15 of t= he DSA) include: designation of contact points for communication with authorit= ies and users; designation of a legal representative in Member States where they are not established; and inclusion in terms of service of information on restrictions applicable to content (European Union, 2022, Arts. 11-15).

Of particular note is t= he obligation for intermediaries to act expeditiously when receiving orders fr= om competent national judicial or administrative authorities to act against specific illegal content or provide information on users. This provision harmonizes, at the European level, the procedures for cooperation between platforms and public authorities.

2.3 Specific obligations for hosting services=

Hosting service providers are subject to additional obligations regarding notice and action. Article 16 of the DSA establishes that providers must implement mechanisms allowing any person or entity to notify the presence of informat= ion they consider to constitute illegal content (European Union, 2022, Art. 16)= .

Notifications must contain specific elements: an explanation of the reasons why the notifier considers the content illegal; a clear indication of the exact electronic location of the content; the name = and email address of the notifier; and a declaration confirming the good faith = of the notification. Providers must process notifications diligently, in a non-arbitrary and objective manner.

Article 17 imposes the obligation to provide reaso= ns for content moderation decisions. When a provider decides to remove or rest= rict access to information, it must communicate to the recipient of the service a clear and specific statement of reasons, including: whether the decision involves removal, disabling access, or demotion; the facts and circumstance= s on which the decision was based; and information on the available avenues of redress (European Union, 2022, Art. 17).

2.4 Obligations for online platforms

Online platforms — defined as hosting service providers that, at the request= of the recipient, store and disseminate information to the public — are subject to additional obligations provided for in Articles 19 to 32 of the = DSA.

Among these obligations, the internal complaint-handling system (Article 20) stands out, which must allow recipie= nts of the service to submit electronic and free-of-charge complaints against content moderation decisions. Complaints must be handled in a timely, non-discriminatory, diligent, and non-arbitrary manner, by appropriately qualified personnel (European Union, 2022, Art. 20).

Article 21 establishes the obligation for platform= s to participate in out-of-court dispute settlement mechanisms certified by nati= onal authorities. Certified bodies must be impartial, independent, and have specialized knowledge.

Article 22 introduces the status of "trusted flaggers," which are entities certified by Digital Services Coordinato= rs due to their expertise and competence in detecting, identifying, and notify= ing illegal content. Notifications from trusted flaggers must be treated with priority (European Union, 2022, Art. 22).

Article 25 imposes specific transparency obligatio= ns for online platform interfaces. The use of "dark p= atterns" — design interfaces that distort or impair users' ability to m= ake free and informed decisions — is prohibited (European Union, 2022, Ar= t. 25).

2.5 Enhanced obligations for VLO= Ps and VLOSEs

The most intensive layer of obligations applies to very large online platforms = and search engines, defined as those reaching at least 45 million monthly active users in the European Union. These providers are subject to the systemic ri= sk assessment and mitigation regime provided for in Articles 34 to 43 of the D= SA.

Article 34 requires VLOPs and VLOSEs to identify, analyze, and assess any systemic risks in the Union stemming from the desig= n, functioning, and use of their services, including: (a) dissemination of ill= egal content; (b) negative effects on the exercise of fundamental rights, particularly dignity, privacy, freedom of expression, and non-discriminatio= n; (c) negative effects on civic discourse, electoral processes, and public security; and (d) negative effects related to gender-based violence, public health, and minors (European Union, 2022, Art. 34).

Article 35 establishes that VLOPs and VLOSEs must implement reasonable, proportionate, and effective mitigation measures tail= ored to the specific systemic risks identified. These measures may include: adaptation of content moderation systems; reinforcement of internal processes; adaptation of terms of service; adapta= tion of algorithmic systems; targeted measures to protect minors; and testing and adaptation of systems through independent auditing (European Union, 2022, A= rt. 35).

The systemic risk regime represents the most innovative aspect of the DSA, reflecting the European legislator's recognit= ion that the largest digital platforms pose specific risks to democracy and fundamental rights that justify enhanced regulatory intervention.

3.1 The intersection between platform regulation a= nd AI

Regulation (EU) 2024/1689, known as the Artificial Intelligence Act or AI Act, complem= ents the DSA by establishing specific transparency obligations for AI systems (European Union, 2024). The intersection between these two instruments is particularly relevant for platform regulation, given the central role that algorithmic systems play in content moderation, recommendation, and curatio= n.

The AI Act adopts a risk-based approach, distinguishing between: (i) prohibited AI pract= ices; (ii) high-risk AI systems; (iii) AI systems with specific transparency requirements; and (iv) general-purpose AI models. For platform regulation purposes, the transparency obligations of Article 50 are particularly relev= ant.

3.2 Transparency obligations for AI-generated cont= ent

Article 50 of the AI Act establishes specific transparency obligations for certain = AI systems. First, providers of AI systems intended to interact directly with natural persons must ensure that those systems a= re designed and developed in such a way that the natural persons are informed = they are interacting with an AI system, unless this is obvious from the circumstances (European Union, 2024, Art. 50(1)).

Second, providers of AI systems that generate synthetic audio, image, video, or text content must ensure that the outputs= of the AI system are marked in a machine-readable format and detectable as artificially generated or manipulated. This obligation is of fundamental importance for combating disinformation and deepfakes (European Union, 2024, Art. 50(2)).

Third, deployers of AI systems that generate or manipulate image, audio, or video content constituting a "deep fake&qu= ot; must disclose that the content has been artificially generated or manipulat= ed. This obligation does not apply when the use is authorized by law for purpos= es of preventing, investigating, or prosecuting criminal offenses, or when the content is clearly artistic, creative, satirical, or fictional (European Un= ion, 2024, Art. 50(4)).

3.3 Interaction between DSA and AI Act<= /span>

The DSA and AI Act are complementary instruments that address different dimensi= ons of platform regulation. While the DSA focuses on the governance of content = and relations between platforms and users, the AI Act addresses the specific ri= sks of artificial intelligence systems, including those used by platforms.=

This complementarity is evident in several aspects. First, both regulations share the transparency principle as a fundamental pillar. Second, VLOPs and VLOSEs that use algorithmic recommendation systems must comply with both the DSA transparency requirements for recommender sys= tems (Article 27 DSA) and the AI Act requirements for AI system providers and deployers.

Third, the use of automated systems for content moderation by platforms is subject to both the DSA rules on statement of reasons (Article 17) and the AI Act requirements, depending on the nature a= nd risk level of the system employed.

The interaction between these instruments creates a comprehensive regulatory framework that addresses the challenges of the dig= ital environment from multiple perspectives, ensuring that platforms are subject= to governance obligations both as content intermediaries and as deployers of AI systems.

&= nbsp;

4.1 The STF decision and the new regulatory paradi= gm

The judgment of Extraordinary Appeals RE 1.037.396 and RE 1.057.258 by the Supr= eme Federal Court on June 26, 2025, represents a turning point for Brazilian platform regulation. The Court declared the partial unconstitutionality of Article 19 of the Marco Civil da Internet, understanding that the general requirement of a prior judicial order for platform liability does not adequately protect constitutional legal interests.

The prevailing thesis established a differentiated regime according to the nature of the illegal content. For serious crimes against constitutional order and human dignity, the Court imposed a "d= uty of care" (dever de cu= idado) on platforms, which implies proactive obligations to prevent and remove such content without waiting for a judicial order. For other types of content, t= he general rule of Article 19 remains applicable.

The STF also established the concept of "syst= emic failure" (falha sist&= ecirc;mica) as a trigger for platform liability. A platform may be held liable when, despite having knowledge or means to know of the widespread presence of cer= tain categories of illegal content, it fails to adopt reasonable measures to pre= vent or mitigate its dissemination. This concept presents functional parallels w= ith the systemic risk regime of the DSA.

4.2 Convergences with the European model

The analysis of the STF decision reveals significant convergences with the Euro= pean regulatory model. The first convergence concerns the graduated approach to platform obligations. The Brazilian model, like the European one, recognizes that not all types of illegal content justify the same regulatory treatment, distinguishing between content that requires proactive action and content subject to the notice-and-takedown regime.

The second convergence relates to the introduction= of proactive duties. The concept of "duty of care" established by the STF approaches the DSA's due diligence logic, requiring platforms to adopt preventive measures proportionate to the state of technology.

The transparency requirement also constitutes a po= int of convergence. The STF determined that providers should establish self-regulation covering a notification system, due process for users, and annual transparency reports. These obligations correspond, in general terms= , to those established by the DSA for online platforms.

4.3 Differences and Brazilian specificities

Notwithstanding the convergences, the Brazilian model presents specificities that must be considered. The first relevant difference concerns the normative source: wh= ile the European regime derives from a regulation with direct and uniform application in all Member States, the Brazilian model emerges from constitutional interpretation by the STF, pending legislative regulation.

The second difference concerns the catalog of seri= ous crimes subject to the duty of care. The STF established an exhaustive list = of criminal types, linked to specific Brazilian legislation, which does not fi= nd direct correspondence in the DSA's systemic risk categories. This option reflects Brazilian constitutional priorities and the context of threats to democracy experienced by the country.

The third relevant difference concerns the absence= , in the Brazilian system, of a formal distinction between large and small platforms. Although the STF recognized the need for proportionality, the European criterion of 45 million users was not transposed to Brazilian law. This gap should be filled by future legislation.

4.4 Perspectives for Brazilian regulation

The STF decision represents an appeal to the legislature to draft new legislati= on capable of remedying the deficiencies of the current regime. In this contex= t, the European experience offers valuable inputs for the Brazilian legislative debate.

From the DSA, elements can be extracted such as: the structure of tiered obligations according= to the size of providers; harmonized notice-and-action mechanisms; the status = of trusted flaggers; transparency obligations for recommender systems; the prohibition of dark patterns in interfaces; and the systemic risk assessment and mitigation regime.

From the AI Act, the following are particularly relevant: the obligations for marking and detecting AI-generated content; t= he requirement to disclose deepfakes; and the transparency requirements for AI systems that interact directly with users. These provisions can contribute = to addressing disinformation and manipulation of public debate through technological means.

The transposition of these elements to the Brazili= an context must, however, observe national constitutional specificities, especially the robust regime for protecting freedom of expression and priva= cy. Brazilian regulation should also consider asymmetries in technological development and the need not to create excessive barriers to innovation and digital entrepreneurship.

CONCLUSION

This article analyzed the European model of digital platform regulation, embodie= d in the Digital Services Act and the AI Act, as a reference framework for the development of Brazilian law after the Supreme Federal Court's decision on Article 19 of the Marco Civil da Internet.

The analysis showed that the European Union has developed, over the past two decades, a sophisticated normative framework f= or the governance of the digital environment, characterized by a tiered approa= ch to obligations, the principle of due diligence, and an emphasis on transparency. The DSA represents a significant evolution from the E-Commerce Directive model, without, however, abandoning the conditional liability exemption regime that constitutes the cornerstone of the system.=

The AI Act complements this regulatory framework by establishing specific transparency obligations for artificial intelligence systems, with particular relevance for synthetic content and deepfakes that challenge the integrity of public debate. The articulation between the two regulations ensures a comprehensive approach to the risks associated with digital platforms in the era of generative AI.

The STF decision of June 26, 2025, brings the Brazilian model closer to the European paradigm by replacing the generalized requirement of a judicial order with a differentiated regime according to t= he nature and gravity of illegal content. The concept of "duty of care&qu= ot; and the notion of "systemic failure" introduced by the Court pres= ent functional parallels with the DSA's systemic risk regime.=

Notwithstanding the convergences, the analysis identified relevant differences that must be considered in the Brazilian legislative debate. The absence of formal criteria for differentiating betw= een large and small platforms, the specificity of the catalog of serious crimes, and the need for infra-constitutional regulation constitute challenges that future legislation must address.

It is concluded that the European experience offers valuable inputs for building the new Brazilian regulatory framework for dig= ital platforms. The structure of tiered obligations, notice-and-action mechanism= s, algorithmic transparency requirements, and the systemic risk assessment reg= ime are elements that can be adapted to the national context, observing constitutional specificities and the country's technological development conditions.

The regulation of digital platforms constitutes a legal field in permanent evolution, requiring legislators and legal practitioners to have the capacity to adapt to technological and social transformations. The dialogue between the Brazilian and European experiences can contribute to the development of regulatory solutions that balance the protection of fundamental rights with the preservation of freedom of expres= sion and innovation, for the benefit of democratic societies.<= /p>

REFERENCES=

BR= AZIL. Law nº 12.965, of April 23, 2014. Establishes principles, guarantees, rights and duties for the use of the Internet in Brazil. Brasília, D= F: Presidency of the Republic, 2014. Available at: https://www.planalto.gov.br/ccivil_03/_ato2011-2014/2014/lei/l12965.htm. Access: Jan. 25, 2026.

BRAZIL. Supreme Federal Court. Extraordinary Appeal 1.037.396/SP<= /b>. Reporting Justice: Min. Dias Toffoli, June 26, 2025. Brasília, DF: S= TF, 2025.

BRAZIL. Supreme Federal Court. Extraordinary Appeal 1.057.258/MG<= /b>. Reporting Justice: Min. Luiz Fux, June 26, 2025. Brasília, DF: STF, = 2025.

BUITEN, Miriam C. The Digital Services Act: from intermediary liabil= ity to platform regulation. JIPITEC - Journal of Intellectual Property, Information Technology and Electronic Commerce Law, [s.l.], v. 12, n. 5, p. 361-380, 2021.

COURT OF JUSTICE OF THE EUROPEAN UNION. Judgment of March 23, 201= 0. Joined Cases C-236/08 to C-238/08 (Google France). ECLI:EU:C: 2010:159.

COURT OF JUSTICE OF THE EUROPEAN UNION. Judgment of July 12, 2011= . Case C-324/09 (L'Oréal v. eBay). ECLI:EU:C: 2011:474.

COURT OF JUSTICE OF THE EUROPEAN UNION. Judgment of May 13, 2014<= /b>. Case C-131/12 (Google Spain). ECLI:EU:C: 2014:317.

EUROPEAN UNION. 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. Off= icial Journal of the European Communities, L 178, p. 1-16, July 17, 2000.

EUROPEAN UNION. Regulation (EU) 2022/2065 of the European Parliament= and of the Council of 19 October 2022 on a Single Market for Digital Services a= nd amending Directive 2000/31/EC (Digital Services Act). Official Journal of the European Union, L 277, p. 1-102, October 27, 2022.

EUROPEAN UNION. Regulation (EU) 2024/1689 of the European Parliament= and of the Council of 13 June 2024 laying down harmonised<= /span> rules on artificial intelligence (Artificial Intelligence Act). Official Journal of the European Union, L, July 12, 2024.

KELLER, Daphne. The right tools: Europe's Intermediary Liability Laws and the EU 2016 General Data Protection Regulation. Berkeley Technology = Law Journal, [s.l.], v. 33, n. 1, p. 287= -364, 2018.

=