As AI sovereignty moves from policy debate to infrastructure reality, Ciena’s Francisco Sant’Anna explores how communications service providers (CSPs) have a pivotal opportunity to play a larger role in enabling regulatory-compliant, high-performance AI ecosystems.

AI has been the defining topic in the networking industry for the past two years, with AI sovereignty increasingly emerging as one of its most consequential aspects. What, until recently, was largely a topic of discussion is now driving significant investments, policy initiatives, challenges, and opportunities across the entire technology ecosystem. Mounting geopolitical instability, combined with AI’s growing economic and strategic importance, has pushed the issue to the top of government and industry agendas.

Data sovereignty, privacy, control, and geopolitics

As AI becomes a central driver of economic growth and national competitiveness, governments are increasingly treating its critical infrastructure—and the networks that connect it—as a national strategic imperative. Sovereignty, once understood primarily in territorial terms, is now extending into the digital domain. It increasingly reflects a nation’s ability to exercise control not only over data and compute, but also over the connectivity that links AI workloads, users (people and machines), models, and data sources.

Countries are adopting policies designed to promote—or in some cases require—local AI infrastructure and control. The aim is to ensure that AI workloads operate under domestic jurisdiction, with data stored and managed locally and models hosted in-country or within trusted regional boundaries.

Beyond geopolitics and economic self-determination, data sovereignty also matters because it underpins the enforceability of national laws and regulations. From data protection and privacy to lawful access and investigative authority, control over where residents’ data is stored, processed, managed, and transmitted helps ensure that domestic legal frameworks apply, rather than those of foreign jurisdictions through which data may traverse or be hosted.

Diverse approaches to AI sovereignty

Regulators, legislators, and governments have been deeply invested in advancing this increasingly important agenda. AI sovereignty is no longer limited to high-level strategy; it is taking shape as infrastructure policy.

Around the world, sovereign AI is taking shape through a range of concrete policy models. The EU is building a coordinated regional compute foundation through its AI Factories and planned AI Gigafactories, while countries such as Canada, India, South Korea, and the UK are advancing sovereign AI through national compute strategies, public-private infrastructure, and directed investment. France offers a related model through cloud and digital sovereignty measures that support AI capability, while Brazil is advancing a sovereign cloud for public-sector data and Singapore is tying AI infrastructure policy directly to power, land, and carbon availability.

These policies have major implications for the networking ecosystem.

The opportunity for network operators

As regional AI infrastructure proliferates, the number of sites requiring high-capacity, low-latency, secure, and policy-aware interconnection is growing rapidly. AI workloads do not operate in isolation; they depend on continuous movement of data between data centers, cloud and edge environments, enterprise users, and distributed data sources. When sovereignty requirements are added to the equation, the network becomes more than a transport layer: it becomes a policy enablement platform.

That shift creates a broader and more strategic role for CSPs, often recognized as national technology leaders. With strong commitments to the regions and communities they serve, and deep relationships across governments, enterprises, and digital infrastructure players, they are well-positioned to participate in sovereign AI ecosystems across four areas:

  • Sovereign connectivity
    CSPs can act as the trusted backbone of sovereign AI ecosystems, providing secure, high-capacity, low-latency, and policy-aware interconnection across AI data centers, cloud environments, edge locations, enterprise sites, and public sector domains. This goes beyond data transport; it includes helping keep traffic within required jurisdictions, connecting localized inference environments, and enabling resilient movement of data across increasingly distributed sovereign AI architectures.
  • AI infrastructure
    Some CSPs may expand beyond connectivity into GPU-as-a-Service, edge AI infrastructure, or direct participation in AI data center builds through ownership models, joint ventures, or strategic partnerships. In the right markets, this can move operators closer to the center of sovereign AI value creation. At the same time, this is a capital-intensive play with meaningful utilization, energy, and operational risks, making it best suited to operators with the right scale, market structure, policy support, and partner ecosystem.
  • Sovereign AI platforms and solutions
    CSPs can also create value by reselling, bundling, hosting, or partnering to deliver AI Software-as-a-Service (SaaS) and Platform-as-a-Service (PaaS) solutions that comply with strict sovereignty requirements. This may include sovereign cloud bundles, regulatory-compliant AI assistants, industry-specific applications, and managed AI platforms tailored to governments and regulated sectors. In many cases, customers will prefer a trusted, integrated offer rather than assembling infrastructure, software, and compliance capabilities separately.
  • Sovereignty assurance and managed operations
    Customers will increasingly need help enforcing jurisdictional, security, governance, and compliance requirements across distributed AI environments. CSPs can play a role in managed interconnection, policy-aware networking, traffic localization, observability, security overlays, and operational assurance—helping customers demonstrate not only AI performance, but also AI management and control.

Not every CSP will pursue all four opportunities, but the path forward is clear: as governments and enterprises seek greater control over where AI runs, how data moves and is managed, and under which jurisdiction digital services operate, the network becomes central to both performance and compliance.

Sovereign AI will not be enabled by compute alone. It will depend on an ecosystem of infrastructure, software, policy, and trust—and the network sits at the center of that system. For network operators, this is not simply a supporting role. It is an opportunity to become a strategic enabler of the next wave of AI infrastructure and service development.

From high-capacity, power-efficient transport to encryption, interoperability, and advanced software automation, Ciena’s technologies and services help CSPs build and operate the scalable, resilient, high-performance networks required to support sovereign AI initiatives.