The Splinternet: Digital Sovereignty and the Geopolitics of Internet Fragmentation

The internet was designed by engineers who believed, with genuine conviction, that a global, open, interoperable network was both technically achievable and politically desirable. The design principles they embedded—end-to-end connectivity, protocol openness, distributed governance, content neutrality—reflected a philosophy that information should flow freely across borders, that no government or corporation should control the network's architecture, and that the connections between people and ideas would, over time, erode the friction that political boundaries imposed.

That vision is not dead, but it is under sustained and escalating pressure. The internet that exists today is increasingly fragmented along geopolitical, regulatory, and commercial fault lines that the original architects did not anticipate and for which the network's technical architecture has no natural defense. States are asserting sovereignty over the digital infrastructure and information flows that cross their borders.

Great powers are building competing technological ecosystems that are increasingly incompatible with each other. Regulatory regimes are diverging in ways that make global compliance either prohibitively complex or strategically implausible. And the commercial interests of platform companies are intersecting with national security priorities in ways that further balkanize the digital landscape.

The result is what some analysts have called the Splinternet—a global network that is fracturing into distinct, partially incompatible zones organized around geopolitical, regulatory, and commercial logics. The process is not sudden or complete, but its trajectory is clear. Understanding where it is heading, what forces are driving it, and what implications it carries for states, corporations, and individuals is one of the defining strategic challenges of the current decade.

The Architecture of Digital Sovereignty

The concept of digital sovereignty is neither new nor monolithic. Different states have invoked it for different purposes, with different institutional architectures and different implications for the global digital ecosystem.

At its most fundamental, digital sovereignty refers to a state's ability to govern the digital infrastructure and information flows within its jurisdiction. This involves several distinct dimensions that are worth disaggregating because they reflect different governance logics and create different kinds of friction with the open internet model.

Data Localization and Jurisdictional Control

The most direct expression of digital sovereignty is data localization—requirements that data generated within a jurisdiction be stored and processed on infrastructure located within that jurisdiction, subject to local law and accessible to local authorities.

Russia's Federal Law No. 242-FZ, enacted in 2014 and enforced from 2015, requires personal data of Russian citizens to be stored on servers located in Russia. China's Cybersecurity Law of 2017 and its Data Security Law of 2021 impose comprehensive data localization requirements on critical information infrastructure operators, with sweeping extraterritorial ambitions. India's Personal Data Protection Bill has contemplated similar requirements. Several EU member states have advocated for data residency within European borders as part of the broader European digital sovereignty agenda, though the EU's approach is more nuanced.

The consequences of data localization for multinational corporations are significant. Operating in multiple jurisdictions with conflicting localization requirements means maintaining separate data infrastructure in each jurisdiction, with associated cost and complexity implications. It also means that global analytics and machine learning applications—which typically require data aggregation across geographies—become more difficult or impossible to operate in the same way, because data cannot be moved freely across borders.

For the structure of the global technology sector, pervasive data localization requirements tend to disadvantage global platforms relative to local operators who can design their infrastructure around a single jurisdiction's requirements from the start. This is, in many cases, the intended effect: digital sovereignty policies are often explicitly designed to create competitive space for domestic technology companies.

Infrastructure Independence and Technological Self-Sufficiency

Beyond data localization, more ambitious expressions of digital sovereignty involve the build-out of domestic technology capabilities that reduce dependence on foreign infrastructure, components, and platforms.

China's strategic investment in domestic semiconductor capability, cloud infrastructure, operating systems, and application platforms reflects a long-term goal of technological self-sufficiency—the ability to operate critical digital infrastructure without dependence on American technology companies, American-controlled standards bodies, or American-controlled hardware supply chains. This effort was significantly accelerated by U.S. export controls on advanced semiconductors, imposed beginning in 2019 and substantially intensified in 2022 and 2023, which demonstrated in concrete terms the vulnerabilities created by technological dependence on a geopolitical rival.

The European Union's approach to digital sovereignty is different in character but reflects similar concerns. The European Chips Act, GAIA-X cloud initiative, and various funding programs for European AI and quantum computing reflect an aspiration for European technological autonomy—not complete self-sufficiency, but a reduced dependence on American and Chinese technology platforms that creates policy leverage and regulatory autonomy. The EU's frustration with the dominance of American cloud providers, and with the privacy implications of American surveillance law's reach over data processed by those providers, has been a consistent driver of European digital sovereignty policy.

Russia's Runet project—the development of a domestic internet infrastructure capable of operating independently if disconnected from the global internet—represents the most extreme version of national infrastructure independence: the ability to maintain domestic digital communications even if the country is cut off from the broader internet. This was tested, partially, during the Ukraine war, and the results illustrated both the technical feasibility of fragmented operation and its practical limitations.

The pursuit of technological self-sufficiency is a response to a genuine strategic vulnerability. Dependence on foreign technology infrastructure creates leverage for adversaries that can be deployed during geopolitical crises. But the costs of self-sufficiency—in innovation foregone, in efficiency lost, in incompatibility introduced—are substantial and often underestimated by the political actors who mandate it.

Content Governance and Information Sovereignty

A third dimension of digital sovereignty involves the governance of information flows—what content is permitted to circulate, what foreign platforms are required to do to operate domestically, and what information about citizens flows to foreign governments.

China's Great Firewall is the most extensive example: a comprehensive system of content filtering, platform blocking, and network inspection that prevents Chinese users from accessing major Western platforms (Google, Facebook, Twitter, YouTube, WhatsApp) while directing traffic to domestic alternatives that operate under Chinese regulatory authority. The result is a largely separate internet ecosystem in which information flows are governed primarily by Chinese regulatory and political priorities.

Russia has progressively tightened content controls since 2012, with significant escalations following the Ukraine war in 2022, when Twitter, Facebook, and Instagram were blocked or restricted and foreign media were designated as "foreign agents" or "undesirable organizations." These measures have not created an equivalent of the Chinese ecosystem—Russians retain significant ability to circumvent restrictions through VPNs—but they represent a direction of travel toward greater content sovereignty.

Even democratic states have moved toward greater content governance, though with different rationales and more limited scope. Germany's NetzDG law requires platforms to remove content that violates German law within specified timeframes. The EU's Digital Services Act imposes content moderation obligations on large platforms. Australia has enacted laws requiring platforms to remove terrorist content. None of these constitute the comprehensive information sovereignty of the Chinese model, but they establish the principle that states have legitimate regulatory authority over what platforms distribute within their territories—and they create compliance requirements that differ across jurisdictions.

The US-China Technology Bifurcation

The most consequential fracture in the global digital ecosystem is the progressive technological decoupling between the United States and China. This decoupling is not simply a consequence of Chinese internet governance choices—it is the product of a deliberate U.S. policy shift, accelerated by the Trump administration but continued and extended by the Biden and subsequent administrations, toward limiting Chinese access to advanced American technology and limiting Chinese technology companies' access to American markets and infrastructure.

The Export Control Escalation

The imposition of U.S. export controls on advanced semiconductors and semiconductor manufacturing equipment—through the Commerce Department's Entity List, the Foreign Direct Product Rule, and the October 2022 and subsequent semiconductor regulations—represents the most significant technology decoupling action in the modern era.

These controls target the specific technologies required to manufacture the most advanced semiconductors: the extreme ultraviolet lithography machines produced by ASML in the Netherlands, the deposition and etch tools produced by American companies like Applied Materials and Lam Research, and the electronic design automation software without which advanced chip design is not possible. By restricting Chinese access to these technologies, the U.S. aims to impede Chinese semiconductor development—and to maintain a technology gap in the military applications of advanced AI that are critically dependent on advanced semiconductor capability.

The geopolitical logic is straightforward: the U.S. security establishment has concluded that unrestricted technology transfer to China in areas relevant to military AI would create unacceptable national security risks. The economic consequences for American technology companies—which had significant China revenues—are treated as acceptable collateral costs of national security objectives.

China's response has been to massively accelerate domestic investment in semiconductor capability. SMIC, China's largest chip manufacturer, has made significant progress in advanced node manufacturing despite the restrictions, though it remains behind the global frontier. Hundreds of billions in state and private investment have been directed toward semiconductor equipment, materials, and design. The trajectory suggests that the restrictions have slowed but not stopped Chinese semiconductor development, and that the gap between Chinese and global frontier capability will narrow over time.

Platform and Application Decoupling

The platform-level decoupling has proceeded separately. Chinese social media platforms—TikTok/Douyin, WeChat, Xiaohongshu—have faced regulatory scrutiny in the United States, with TikTok subject to legislation requiring its sale or ban. American platforms remain largely blocked in China, with the partial exception of LinkedIn (which has since withdrawn from China's social platform business) and limited access for enterprise applications.

This platform bifurcation has significant implications for the global information ecosystem. The roughly 1.4 billion internet users in China live in a largely separate information environment from the 3+ billion users in the rest of the world. Content that goes viral in one ecosystem often has no presence in the other. Information narratives that shape public opinion in one zone are invisible in the other. This is not merely a commercial phenomenon—it has implications for how different populations understand geopolitical events, scientific developments, and global challenges.

The platform bifurcation is also creating divergent development trajectories. Chinese AI companies, cut off from some Western training data and model weights, are developing their own foundational AI capabilities. American and European AI companies face restrictions in deploying their most capable models in China. The result is the emergence of parallel AI ecosystems with different technical architectures, training data, and capability profiles—a divergence that will compound over time.

The Semiconductor Geopolitics

No domain better illustrates the intersection of technology and geopolitics than semiconductors—the physical substrate on which the digital economy and modern military capability are built. Control over the semiconductor supply chain has become a primary site of strategic competition between the United States and China, with implications that extend well beyond bilateral relations.

The semiconductor supply chain is among the most complex and globally distributed in the world. Design tools originate primarily in the United States (Cadence, Synopsys, Mentor). Chip design is concentrated in the U.S. (Nvidia, AMD, Qualcomm, Apple) and Taiwan (MediaTek). Advanced manufacturing is dominated by TSMC in Taiwan and Samsung in South Korea. Equipment for manufacturing comes from the U.S. (Applied Materials, Lam Research, KLA), the Netherlands (ASML), and Japan (Tokyo Electron, Shin-Etsu Chemical). Raw materials come from Germany, Japan, and increasingly China.

This globally distributed supply chain creates strategic vulnerabilities for every major power. For the United States, the concentration of advanced manufacturing in Taiwan—a territory China claims and has threatened to reunify by force—is an existential strategic risk. For China, dependence on American design tools, Dutch lithography equipment, and Taiwanese fabrication creates leverage that adversaries can exploit. For Europe, dependence on both American and Asian components across the chain creates both risk and policy complexity.

The consequence is a global race to build more national semiconductor capability: the U.S. CHIPS Act, the EU Chips Act, India's semiconductor incentive programs, Japan's subsidized partnership with TSMC, and China's Semiconductor Big Fund. This industrial policy scramble will not—and cannot—produce equivalent semiconductor capabilities in every jurisdiction, but it will produce more geographically distributed production, more redundancy in critical nodes, and more national capacity to apply political pressure through supply chain leverage.

The European Third Path

The European Union occupies a distinctive position in the geopolitics of digital fragmentation—neither aligned with the maximally open American model nor with the maximally sovereign Chinese model, but pursuing a third path that emphasizes regulatory sovereignty, fundamental rights protection, and strategic autonomy.

The Regulatory Sovereignty Approach

The EU's approach to digital governance is grounded in the assertion that European institutions have the authority and the obligation to regulate digital markets and digital rights within European jurisdiction—even when the entities being regulated are primarily American or Asian technology companies. This assertion, backed by large fines and binding behavioral requirements, has established the EU as the world's most consequential digital regulator.

The GDPR, enforced from 2018, established comprehensive privacy rights for EU residents and imposed compliance obligations on any company that processes personal data of EU residents, regardless of where the company is located. Its extraterritorial reach, backed by fines of up to 4 percent of global annual turnover, has shaped global privacy practices. Many American and Asian companies have adopted GDPR-compliant practices globally, because the compliance cost of maintaining separate privacy frameworks for EU and non-EU users exceeds the cost of uniform global compliance.

The Digital Markets Act, entering full enforcement in 2024, designates specific "gatekeeping" platforms—Apple, Google, Meta, Amazon, Microsoft, TikTok—and imposes specific interoperability, data access, and behavioral requirements on those platforms in their European operations. The Digital Services Act establishes comprehensive obligations around harmful content, algorithmic transparency, and advertising practices. The AI Act, the world's first comprehensive AI regulatory framework, imposes risk-based requirements on AI systems deployed in the EU.

Collectively, these regulations assert a comprehensive European jurisdiction over the digital sphere—not by building Chinese-style firewalls but by establishing regulatory requirements that platforms must comply with to operate in Europe. This creates compliance complexity for global platforms and divergence between European and non-European digital markets, but preserves the basic connectivity of the internet within and through Europe.

The Sovereignty Dilemma

The EU's digital sovereignty agenda contains an internal tension that has not been fully resolved. European leaders want both to protect European digital rights and to reduce dependence on American and Chinese technology platforms—but these goals can conflict.

The GDPR's privacy protections were largely designed in response to American surveillance law: the concern that American cloud providers, subject to intelligence community requests under FISA Section 702 and similar authorities, were providing U.S. intelligence agencies with access to European personal data. The Schrems I and Schrems II rulings by the European Court of Justice invalidated successive EU-U.S. data transfer frameworks on exactly this basis.

Yet the alternative to European data being processed by American cloud providers—European data being processed by European cloud providers—requires European cloud infrastructure to exist at adequate scale and quality. The current generation of European cloud providers (Deutsche Telekom's Open Telekom Cloud, Thales's S3NS, OVHcloud) are substantially smaller and less capable than American hyperscalers. GAIA-X, the Franco-German initiative to build a European cloud ecosystem, has produced governance frameworks but limited actual infrastructure.

The European digital sovereignty project faces the uncomfortable reality that genuine technological autonomy requires technological capability, and building that capability requires sustained investment over decades. The regulatory tools Europe has deployed are effective at imposing compliance obligations on existing platforms; they are less effective at building new ones. The gap between regulatory ambition and industrial capability remains significant.

European Strategic Autonomy in Practice

Despite these challenges, the European approach has produced real strategic effects. The EU's regulatory posture has given it genuine leverage over global technology companies—leverage that has been deployed to extract behavioral changes, significant fines, and, in some cases, structural modifications. European companies operating in digital markets have more privacy protection, more algorithmic transparency, and more platform interoperability than users in most other jurisdictions.

The EU has also positioned itself as a template for digital governance for other jurisdictions—the "Brussels Effect" phenomenon in which EU regulatory standards become de facto global standards because the cost of maintaining separate compliance frameworks is too high. GDPR-like privacy laws have been adopted or are being developed in many jurisdictions outside Europe, partly reflecting genuine convergence of values and partly reflecting the practical dominance of EU standards.

Implications for Multinational Corporations

For businesses operating across the digitally fragmenting global landscape, the strategic implications are profound and concrete.

Compliance Complexity and Architectural Divergence

The practical compliance burden created by diverging digital regulatory regimes is substantial. A global technology company operating in China, the European Union, and the United States faces requirements that are in some respects mutually incompatible:

• Chinese data localization requires that Chinese user data be stored in China and accessible to Chinese authorities.
• EU GDPR requires that personal data be protected against access by foreign governments without adequate legal protections—including potentially Chinese and American authorities.
• American law provides for government access to data held by American companies globally under various legal authorities.

Complying simultaneously with all three regimes while serving users in all three jurisdictions requires architectural choices—separate infrastructure, data siloing, jurisdictional firewalls within the company's own systems—that are costly and operationally complex.

Companies that handle this well tend to have developed explicit data architecture strategies that map compliance requirements to system design choices from the outset, rather than retrofitting compliance onto systems designed for a unified global architecture. This requires involving legal, compliance, and regulatory expertise in fundamental architectural decisions—not as reviewers after the fact but as contributors to the design process.

Market Access Trade-offs

The fragmentation of digital infrastructure also forces market access choices that previously could be deferred. Operating in China typically requires commitments to data localization, content moderation standards, and government access arrangements that may conflict with the company's policies in other jurisdictions and with its home country's regulatory requirements.

American technology companies have navigated this tension in different ways. Many have exited or significantly reduced their China operations—Google, Facebook, Twitter/X, LinkedIn (in its social media product). Others maintain China operations through locally incorporated joint ventures or subsidiaries with substantial operational independence, enabling compliance with Chinese requirements while limiting the contamination of global operations. A few maintain genuinely integrated China operations with the associated compliance complexity.

The trade-off is not simply commercial. Market access decisions in politically sensitive contexts carry reputational implications with employees, customers, regulators, and investors in home jurisdictions. The question "what concessions to local requirements are acceptable for market access?" has become a governance-level strategic question for boards and executive leadership—not merely an operational trade-off.

The Localization Premium and Competitive Dynamics

Digital fragmentation systematically advantages local competitors over global ones in markets where digital sovereignty requirements are most demanding. A domestic Chinese technology company does not face the compliance complexity of operating under both Chinese and international regulatory frameworks—it is simply Chinese, operating under Chinese law, with Chinese infrastructure.

This localization premium has produced distinct national technology champions in several jurisdictions: Baidu, Alibaba, WeChat, and ByteDance in China; Yandex in Russia; Kakao and Naver in South Korea; various European fintech and cloud providers. These companies have competitive advantages in their home markets that derive not just from local knowledge but from regulatory positioning that global competitors cannot fully replicate.

For multinationals, the strategic response involves several choices: accepting the localization premium and competing on quality and differentiation rather than regulatory parity; building local entities with genuine operational independence that can compete as quasi-local players; or focusing on the markets and segments where global scale and integration create advantages that local players cannot match.

Internet Standards and Technical Fragmentation

Beyond regulatory and commercial fragmentation, there are emerging concerns about technical fragmentation—divergence in the protocols, standards, and architectures that define the internet itself.

The internet's current architecture depends on shared standards—TCP/IP, HTTP, DNS, TLS—that are developed and maintained through international standards bodies (IETF, W3C, IEEE) operating through rough consensus processes. These bodies have historically been dominated by American and Western European technologists and reflect technical values—openness, interoperability, decentralization—that align with Western internet governance philosophy.

China and Russia have advocated in international forums, particularly the ITU (International Telecommunication Union), for alternative internet governance models that would give national governments more explicit control over their portions of the internet. These proposals have not succeeded in changing the fundamental technical architecture of the internet, but they reflect a sustained effort to shift governance authority from the current multi-stakeholder model toward a more intergovernmental framework.

The technical protocols of the internet are among the few truly global commons remaining in a fragmenting geopolitical environment. Their governance through multi-stakeholder bodies represents a distinctive institutional achievement that is increasingly under pressure. The ability to maintain functional global standards in the face of geopolitical fragmentation is not guaranteed—and its loss would be more consequential than most current discussions acknowledge.

5G and the Telecommunications Architecture

The geopolitics of 5G telecommunications infrastructure has become one of the clearest examples of technology fragmentation driven by security concerns. American pressure on allied governments to exclude Huawei and ZTE equipment from their 5G networks—driven by concerns about Chinese intelligence access to network infrastructure—has produced a distinct split in global telecommunications architecture.

Most NATO members and U.S. allies have restricted or excluded Chinese 5G equipment, relying instead on European vendors (Ericsson, Nokia) and, to a lesser extent, Samsung. China and countries less aligned with the U.S. security community have deployed Huawei infrastructure extensively. The result is a global 5G architecture that is divided along geopolitical lines—not technically incompatible, but creating distinct supply chains, vendor relationships, and potential intelligence access profiles.

The 6G development process will face the same dynamics at a critical juncture. The standards for 6G telecommunications are being developed now, and their architecture will shape telecommunications infrastructure for the following decade. The contest over 6G standards—in which Chinese companies are highly active participants, and in which the U.S. has sought to limit Chinese influence—is one of the defining technical governance battles of the current period.

AI Standards and Governance Fragmentation

Artificial intelligence is becoming a new frontier of standards and governance fragmentation. The EU's AI Act establishes a risk-based regulatory framework for AI systems deployed in Europe. China has enacted its own AI governance regulations—including rules on generative AI, algorithmic recommendation systems, and deep synthesis—that reflect Chinese regulatory priorities. The United States has pursued a lighter-touch approach, with executive orders and voluntary commitments rather than comprehensive legislation, though this may evolve.

The divergence in AI governance frameworks creates compliance complexity for AI companies operating globally—different disclosure requirements, different risk classification systems, different prohibited use categories. More fundamentally, it reflects different value priorities: the EU emphasizes fundamental rights protection and human oversight; China emphasizes stability and social harmony; the U.S. has thus far emphasized innovation and market development.

The absence of international AI governance coordination—despite several multilateral efforts including the G7 Hiroshima AI Process and the AI Safety Summit at Bletchley Park—means that the regulatory divergence is likely to persist and may deepen, particularly as AI systems become more capable and more consequential.

The Developing World and the Architecture of Digital Dependence

Much of the discussion of digital fragmentation focuses on the triangular competition among the United States, China, and Europe. But the most consequential choices about digital architecture may be made by the roughly 100 countries that are neither major powers nor regional technology leaders.

These countries face a genuine choice architecture in their digital infrastructure decisions: whether to build on American platforms and infrastructure (with associated dependencies on American law, policy, and corporate governance), on Chinese platforms and infrastructure (with associated dependencies on Chinese government access requirements and Belt and Road financing terms), or on European frameworks (with associated compliance obligations and regulatory requirements). Many lack the resources to build genuine national alternatives.

China has been particularly systematic in offering digital infrastructure investment through the Digital Silk Road component of the Belt and Road Initiative—submarine cables, data centers, mobile networks, and smart city infrastructure financed and often built by Chinese companies under terms that typically include provisions for data access and network management that raise security concerns among China's geopolitical adversaries.

The pattern of digital infrastructure investment tracks, and in some cases anticipates, geopolitical alignment. Countries that have accepted significant Chinese digital infrastructure investment have moved toward greater operational dependence on Chinese technology companies and greater exposure to Chinese government access to their network infrastructure. This creates a mechanism for expanding Chinese geopolitical influence that operates through commercial relationships rather than explicit coercion.

The architecture of digital infrastructure is an architecture of political dependence. The countries that control the cables, the servers, the software platforms, and the standards have leverage over those that use them—leverage that can be deployed economically, commercially, and in extremis as a geopolitical tool. The strategic significance of which companies build the world's digital infrastructure has been poorly understood until recently; it is now one of the central strategic contests of the era.

Corporate Strategic Implications

For senior business leaders, the fragmentation of the global digital ecosystem is not an abstract geopolitical phenomenon—it is a concrete operating environment that demands explicit strategic responses.

Digital Architecture as Strategic Choice

The architecture of a company's digital infrastructure—which cloud providers it uses, where its data is stored and processed, how its applications are designed across jurisdictions—has become a strategic choice with geopolitical implications. Companies that built their digital architecture on the assumption of a unified, globally accessible internet are now having to retrofit compliance structures onto systems not designed for fragmentation.

The strategic imperative is to bring the same rigor to digital architecture decisions that leading companies bring to supply chain decisions. Where are our critical data assets? What jurisdictions govern their storage and processing? What government access requirements apply? Where are the single points of failure in our digital supply chain? These questions should be addressed at the board level, with explicit governance responsibility assigned.

Market Prioritization in a Fragmenting Digital World

Digital fragmentation forces more explicit choices about which markets to prioritize and how to operate in each. For most companies, attempting to maintain equivalent digital capabilities across China, the EU, and the U.S. with full regulatory compliance in each is not feasible—the requirements are too divergent and the compliance costs too high.

More realistic is a portfolio approach: deciding which markets to operate in with full investment, which to serve through partnerships or licensing arrangements that shift compliance responsibility to local partners, and which to exit because the compliance burden exceeds the commercial opportunity. This portfolio decision should be made explicitly, at the strategic level, with clear-eyed assessment of regulatory requirements, commercial opportunities, and reputational implications.

Investing in Digital Policy Capability

The fragmentation of the global digital regulatory environment has increased the value of sophisticated digital policy capability—the organizational ability to engage with regulators, influence standards processes, and navigate complex compliance environments across jurisdictions. Companies with weak digital policy capability are increasingly disadvantaged relative to those with experienced teams able to engage substantively with the EU's DG Connect, China's MIIT, or the FTC in the United States.

This capability is not primarily about legal compliance—though legal compliance is necessary. It is about the ability to shape the regulatory environment through active engagement: providing technical expertise to regulators, participating in standards bodies, contributing to policy consultations, and building relationships with regulatory staff before enforcement becomes an issue.

Building Resilience Against Digital Coercion

The experience of companies that have had operations disrupted by digital coercion—data demands that conflict with home country law, infrastructure dependencies that create leverage for government pressure, market access conditions that require unacceptable concessions—provides lessons for building resilience.

Key elements of digital resilience include: data architecture that avoids single points of jurisdictional control over critical data assets; vendor diversification that reduces dependence on any single supplier whose access might be restricted; operational designs that can continue to function if specific jurisdictions become inaccessible; and board-level governance clarity about what concessions are acceptable under pressure.

The Future Architecture of a Fragmented Internet

The direction of travel in digital fragmentation is relatively clear, even if the ultimate destination remains uncertain. Several structural forces are driving fragmentation: geopolitical competition between great powers, regulatory sovereignty assertions by states across the political spectrum, and the commercial interests of both national champions and global platforms.

The question is not whether the internet will fragment but how far the fragmentation will go. Three scenarios are worth considering.

Managed fragmentation envisions a world where the internet remains operationally interconnected but is governed through a patchwork of national and regional regulatory frameworks that create compliance complexity without preventing connectivity. Users in different jurisdictions see different content, operate under different privacy regimes, and interact with different platform ecosystems—but the fundamental technical infrastructure remains interoperable. This is roughly the current trajectory.

Technical bifurcation envisions a more profound split in which the U.S./European and Chinese digital ecosystems become genuinely technically incompatible—different protocol stacks, different AI architectures, different identity and authentication systems. This outcome would require more deliberate decoupling than has occurred so far, but the building blocks are being laid.

Comprehensive national internets represents the extreme endpoint: a world in which each major national market operates a largely self-contained national internet, with the global internet reduced to a set of interconnect agreements among national networks subject to granular government control. Russia's Runet project aspires to this model; China's Great Firewall is its closest current realization.

The costs of advancing digital fragmentation fall unevenly. They are borne primarily by users in smaller, less technically capable countries who lose access to the full breadth of global information and services; by multinational companies that must maintain multiply-redundant digital operations; and by the global science and research community, whose cross-border collaboration depends on open information flows. They are paid primarily by national governments and technology incumbents whose interests are served by balkanization.

The actors with the most ability to influence the trajectory are the major powers—particularly the United States and the European Union, whose combined regulatory authority, market size, and technology leadership give them significant influence over global digital architecture choices. The current trajectory is driven by competitive pressures that neither power fully controls but that both could, with sufficient political will and coordination, moderate.

The window for moderating digital fragmentation is narrowing. Technical investments in national infrastructure, standards bodies positions, and regulatory frameworks are creating path dependencies that become more difficult to reverse over time. Decisions made in the next five years about digital infrastructure architecture, AI governance frameworks, and semiconductor supply chains will shape the digital landscape for decades.

This is not merely a policy question. It is a strategic question for every organization operating in the global economy—about where to invest, how to structure operations, which regulatory engagements to prioritize, and how to build resilience against a digital environment that is becoming less predictable, less open, and more politically contested than the one for which most organizations' digital strategies were designed.

---

Sources & References

• Freedom House, Freedom on the Net (annual report)
• Council on Foreign Relations research on internet fragmentation and digital sovereignty
• European Commission official communications on Digital Sovereignty and the EU Digital Decade
• China's Cybersecurity Law, Data Security Law, and Personal Information Protection Law (official texts and analysis)
• Brookings Institution research on U.S.-China technology decoupling
• MIT Technology Review coverage of semiconductor geopolitics
• McKinsey Global Institute research on data flows and digital fragmentation
• RAND Corporation research on digital authoritarian diffusion
• CSIS (Center for Strategic and International Studies) research on 5G and telecommunications security
• OECD reports on data localisation and cross-border data flows
• Stanford Internet Observatory research on information operations and platform governance
• Electronic Frontier Foundation research on global internet governance
• Journal of International Affairs and Foreign Affairs (selected articles on digital sovereignty)
• Atlantic Council research on the Digital Silk Road and Chinese technology investment
• European Council on Foreign Relations research on European digital strategic autonomy
• Chatham House research on internet governance and geopolitical fragmentation
• ICT4Peace Foundation research on international cyber norms
• Internet Society annual reports on the state of the internet
• ITU Global ICT Development Index and related reports