Chapter 7 builds on the analysis developed in the previous chapters and starts from the premise that, from a strictly technological perspective, AI reshapes entrepreneurial activity without undermining the entrepreneur’s fundamental role within capitalism. The chapter argues that the most significant risks associated with AI-driven transformations do not arise from technology itself, but rather from a set of human dynamics. In particular, it highlights the growing dominance of large AI giants and the emergence of forms of systemic lock-in that extend across science, talent, venture capital, acquisitions, and inferential infrastructure. These dynamics constrain competitive conditions and may significantly limit the scope for start-ups to achieve independent and sustainable success. The analysis further considers additional sources of pressure stemming from broader macroeconomic and political developments, as well as from sociological processes such as algorithmic polarization and homogenization, and the rise of ‘socially thin innovation’, in which novelty is increasingly detached from substantive social value. Taken together, these forces risk encouraging multiple postmodern forms of entrepreneurship that remain confined within the boundaries set by dominant platforms. The chapter concludes by characterizing this emerging configuration as a form of ‘fenced capitalism’, in which the space for genuinely disruptive innovative entrepreneurship becomes progressively entrapped.

Chapter 6 examines the start-up landscape in the AI era. It proposes a taxonomy of ventures based on whether AI represents the firm’s core product or, instead, functions primarily as an enabling tool, and it further distinguishes ventures by the underlying AI capability, from predictive to generative (and increasingly agentic) systems. The chapter identifies four archetypes – predictive analytics ventures, creativity-centric generative AI start-ups, business intelligence-enhanced AI-served start-ups, and creativity-augmented AI-served start-ups – while emphasizing that the boundaries between these categories are becoming progressively porous, especially as firms combine multiple functions and capabilities. The chapter highlights contrasting competitive conditions across domains: incumbents benefit disproportionately from proprietary data in predictive applications, whereas generative AI tends to lower entry barriers and widen the pool of potential entrants. It then distils three stylized facts: the tight coupling between AI science and entrepreneurial activity, the emergence of a ‘talent trap’ that draws top AI engineers towards large high-tech firms, and the pervasive influence of MAMAA and other AI giants over innovative AI-based start-ups. The chapter concludes by suggesting a cultural shift in which entrepreneurial success is increasingly evaluated through acquisition rather than independent growth, with significant social welfare implications.

The widely accepted “AI Race” narrative, which pits the US against rivals like China, is a flawed metaphor that actively harms thoughtful AI development and policy. This chapter deconstructs the “race” by showing it lacks a clear finish line or coherent definition of “winning,” unlike historical contests such as the space race or the Manhattan Project. The narrative is based on three shaky assumptions: that innovation inherently equals progress, that generative AI is already capable of near-term, large-scale usefulness, and that the “winner” will control global AI values. Critically, these flawed premises trigger predictable, problematic policy reactions, including unlimited corporate and governmental spending, indefinite delays on necessary regulation, and hasty, thoughtless deployment. By prioritizing speed and raw capability over genuine utility, the “AI race” risks squandering resources on unreliable systems and undermining public trust. The chapter concludes that the real challenge lies in human coordination, not computation, and calls for abandoning the race framework in favor of a wise, goal-oriented approach that serves human needs over corporate interests.

This chapter challenges the popular narrative that open-source software will democratize generative AI (GenAI) and disrupt the existing oligopoly dominated by companies like Google, Microsoft, Meta, and OpenAI. It argues that Big Tech routinely “weaponizes” open-source projects (e.g., Linux, Android, PyTorch) as a strategic tool to gain a competitive advantage over rivals, not for true democratization. The analysis shows that despite impressive percentage growth from startups, the absolute market value gains of tech giants have dwarfed those of new entrants. The enduring concentration of power stems from eight key levers favoring Big Tech: vast capital, exclusive proprietary datasets, massive access to compute (GPUs), network effects, control over top talent, market access agreements, a high-interest investment environment, and a lack of a viable business model to unseat their integrated ecosystems. Ultimately, the chapter concludes that open source in GenAI acts as a “fig leaf” for corporate dominance, as these massive companies control nearly every link in the GenAI supply chain – making radical openness unlikely to have a meaningful impact on the existing power structure.

This chapter discusses the gradual consolidation of the dual-sided framework in the contemporary digital era. Despite the decades-long transatlantic battle between ‘free flow’ and privacy protection, serious disagreement over how to conceptualize data, and the inter-regime tension between human rights and world trade, this chapter argues that the consolidation of the dual-sided framework is precisely a result of such internal tensions and contradictions. The more this dichotomy seems hard to resolve, the more the two framings of free trade and human rights get reinforced, and the more they monopolize the policy terrain for the governance of information and data. This chapter also shows that, despite the entrenched dichotomy, free trade and human rights regimes have been mutually influencing in the ways they respectively approach transborder data flows and the responsibility of big tech, hence further reinforcing the transnationalization of capital and unequal distribution of power in information capitalism.

By 2022, social media platforms had become more prominent access points to news than traditional news media platforms. Although news media also draw benefits from online platforms, they find themselves in an increasingly asymmetric relationship that appears to harm journalists and new media’s ability to generate revenues online. Remedying the uneven playing field between big tech and news media has been a recurring ambition of European media policy. In the context of literature on corporate political activity, this chapter investigates how three big tech companies, Alphabet, Meta and Microsoft, have positioned themselves in relation to EU policymaking that aims to strengthen the rights of journalists, news media and press publishers online. The chapter argues that while these big tech companies primarily seek to preserve their own business models and reputations, the media policy domain also reveals a split in their lobby narratives. Due to lower exposure to reforms in digital media policy, Microsoft has been less opposed to, and in fact has campaigned for, stronger protection for news media against digital platforms.

Chapter 5 focuses on the regulation of social media platforms and platform architecture, with changes in EU perceptions regarding the reliability of these platforms and the values of their owners. It examines the shift from economically motivated self-regulatory regimes in these sectors based in logics of efficiency to a digital sovereignty-motivated move to a logic of security in regulation. It identifies the explicit linkage between economic and security concerns, particularly as it relates to disinformation and political advertising, with the promotion of co-regulatory regimes with significant levels of oversight provided by the Commission. It explores the approach to regulatory export adopted in these initiatives, with an emphasis on control of platforms regardless of where they are based, so long as they offer services in the EU.

This chapter explores the prospects of Pillar One and Pillar Two. Fundamentally, Pillar One seems destined to fail, while Pillar Two is being implemented by a critical mass of countries. Pillar One is unlikely to succeed for three reasons. First, it requires a multilateral tax convention (MTC) among over 100 countries to be implemented. Second, because Pillar One is still aimed primarily at taxing the US digital giants (Big Tech), it is hard to envisage it being implemented without the United States. Third, Pillar One is premised on all the countries that have adopted DSTs repealing them. But DSTs are politically popular and, in some cases (e.g., the United Kingdom), brought in significant revenue. This chapter discusses options countries may have to tax Big Tech without Pillar One and then addresses the US response. For Pillar Two, it examines how Pillar Two will affect countries that have not adopted it yet, such as China and the United States, and Pillar Two’s impact on domestic tax incentives like tax holidays, accelerated depreciation, and tax credits.

The Introduction sets the scene by describing the relevance of the idea of a symbiotic relationship between competition and democracy in the history of competition law and the contemporary policy debate. It identifies the gap in the literature, explains the research question and the purpose of the book, and presents the argument of the book in a nutshell. It starts with the observation that the idea of a link between competition and democracy is a recurrent theme in US and EU competition law. However, existing scholarship has so far struggled to clearly explain what the relationship between competition and democracy actually consists of. This knowledge gap is filled by this book which provides a clear, conceptually sound, and surprising answer to this research question. It argues that the idea of competition–democracy nexus is grounded in a republican understanding of liberty as non-domination which can be traced back to the political thought of the Ancient Roman republic and fundamentally differs from our contemporary negative concept of liberty as non-interference. The purpose of the book is to demonstrate how this republican concept of liberty explains the idea of a competition–democracy nexus in US and EU competition law.

Datafication-enabled advertising and other datafication practices, in the absence of proper constraints, will deepen the perils of datafication. A set of cross-border competition disciplines proposed in Chapter 5 may well be an effective instrument to address problems associated with platform monopolies and data capitalism. In this way, there would be less need for ex-post competition law enforcement in developing countries and LDCs, where relatively limited resources can be allocated to combat digital cartels and data monopolization. In this context, algorithmic transparency can serve as a starting point for global platform governance. The case study in Chapter 5 investigates the key dimensions of platform transparency requirements in a comparative context and demonstrates that the fragmentation of platform regulation is growing. The proliferation of platform regulations and algorithmic disciplines may place SMEs in an even more difficult situation vis-à-vis big tech companies, which have the resources necessary to manage different legal requirements in different countries. Despite the inherent complexity of the political economy surrounding digital capitalism, Chapter 5 concludes that there are reasons to be optimistic about better governance through international trade agreements.

Media platformization has caused problems that a state cannot easily regulate. The media content regulations considered in Chapter 4 are prime examples demonstrating the need to alter the power distribution in the Internet ecosystem. In terms of speech platforms, the “rules-based” platform governance model led by the EU, such as the DSA, now represents a strong power in balancing the US’ CDA-based social media self-regulation. The years to come will be critical to the global governance of digital platforms. Key indicators include whether more and more countries will adopt DSA-like regulations along the EU regulatory path. In terms of streaming services, the content quotas for video streaming platforms may constitute performance requirements for investment in services. This measure may also violate the obligations that apply to the nondiscriminatory treatment of digital products. Looking ahead, cultural diversity concerns regarding avatars in the VR space will be even more complex and will propel the “trade v. culture” clash to another level. No matter how carefully crafted, media content regulation must face enforceable reality. Trade rules that ban local presence would enable platform companies to supply services without establishing a local presence, which could significantly constrain a state’s ability to enforce platform regulations.

So as not to face potential disruption, universities will need to actively engage in the sector’s evolution as opposed to passively observing its revolution. To stay competitive, they will need to adapt pedagogy and content to ongoing changes lest they risk their irrelevance. They are advised to create occasions for relationship and community building, fostering students’ affection for and attachment to their alma mater, and to avoid the university’s being abstract in students’ minds. Instead of the current standard of one-time, early-life degrees (intermittence), lifelong learning and persistence over alumni’s entire professional careers appear to be academia’s model for the future. Universities must assure higher applicability of taught material to job requirements, as employers are increasingly launching own corporate universities offering nano- and micro-degrees. Finally, affiliation with edtech start-ups and big tech companies might be necessary to ensure the funding needed to navigate academia’s new online reality and thus bypass isolation.