- What should lagging countries do before they become technologically dependent?
Delays in adopting artificial intelligence and digital sovereignty are no longer a developmental matter that can be postponed; it has become a matter of national security and a state's ability to sustain itself. The ongoing wars, from Ukraine to Gaza, through the Red Sea disturbances and the regional escalation that has affected energy markets and supply chains, have proven that a state that does not own its data, does not control its digital infrastructure, and cannot operate its vital services during outages, is a state exposed to losing its decision-making power before losing its resources. Meanwhile, World Bank reports indicate that developing countries that do not build their AI foundations now will face a wider gap in productivity and competitive ability because AI relies on four intertwined pillars: connectivity, computing power, data, and skills.
The problem is that many lagging countries still treat AI as a pilot program or publicity path, while major powers treat it as a sovereign infrastructure including data centers, energy, network security, semiconductor chains, data control, regulation, skills, and the capacity to operate in wartime and crises. This shift is also clear in international discussions about “digital sovereignty” and “AI sovereignty,” where the issue is no longer just about who buys the best model, but who has the local or regional capability to operate the models, protect the data, and maintain vital services during interruptions or blackmail or sanctions or supply chain disruptions.
The first lesson from current wars: Infrastructure is no longer just operational background but a front of conflict
The most important revelation from the war in Ukraine is that the infrastructure of energy, telecommunications, and civil systems is not merely a collateral target, but part of the battlefield itself. The International Energy Agency concluded, based on the Ukrainian experience, that resilience must become a focal point of energy planning, with the need to fortify vital assets, decentralization, critical equipment storage, and considering data as a strategic asset, while integrating cyber resilience at all planning and operating stages. The World Health Organization documented that attacks on health care in Ukraine increased by about 20% in 2025 compared to 2024, and attacks since February 2022 have affected thousands of facilities and staff, illustrating how quickly a collapse in electricity or transportation or communications can lead to a direct health collapse.
The strategic outcome here is clear: AI strategies can no longer be separated from business continuity plans, crisis management, civil defense, and national resilience. A state that builds a smart model over a fragile network or unstable energy or external cloud infrastructure is essentially building a layer of intelligence over shaky ground.
The second lesson: Digital sovereignty is not isolation, but the ability to make decisions and control
Reducing digital sovereignty to “localising everything” or “preventing data flow” is mistaken. OECD reports affirm that the real challenge is to build data flows across borders with confidence; that is, the ability to leverage the global digital economy while ensuring oversight, protection, compliance, and accountability. In other words, sovereignty does not mean isolation from the world, but rather that the state has the legal and technical and operational ability to know where its data is, who accesses it, how it is processed, and what happens if the service is interrupted or the supplier changes or a political or cyber dispute arises.
Therefore, lagging countries do not need an ideological discourse on sovereignty, but a sovereign architectural design: locally or regionally sensitive data, less sensitive hybrid services, national key management and encryption, interoperable operating standards, and a real ability to transfer loads from one provider to another. Sovereignty here is not about “owning everything,” but preventing dependence on any single point of failure.
The third lesson: AI in wartime begins with energy and computing before algorithms
The World Bank summarized the basics of AI readiness in the “four Cs”: Connectivity, Compute, Context, Competency; that is, connection, computing, the data context, and skills.
