The European Bank for Reconstruction and Development has warned that many economies across its regions risk being left behind as global investment in artificial intelligence and defence accelerates. In its Transition Report 2026–27, published on Tuesday, November 17, the bank says gaps in electricity supply, digital infrastructure, financing and industrial capacity could prevent countries from turning rising demand into new exports and higher productivity.

A widening gap in the AI supply chain
The report, titled Frontiers of Competitiveness, examines how technological change and shifting geopolitical priorities are reshaping demand. The EBRD works across about 40 economies in Eastern Europe, Central Asia, the Middle East and Africa. Its assessment is therefore focused on a broad and diverse group of countries rather than on one national market.
The bank says economies in its regions account for about 8 per cent of global merchandise exports overall but only 0.8 per cent of exports of AI-related products. Advanced chips, large-scale data infrastructure and frontier AI models remain concentrated in a small number of countries and companies. The report also identifies electricity constraints: in almost half of the EBRD economies, existing power grids would not be able to support even one data centre.
That infrastructure gap matters because AI development and deployment depend on reliable electricity, high-capacity connectivity, computing resources, skilled workers and access to capital. Without those foundations, countries may be able to use imported AI tools but struggle to host the infrastructure, firms and high-value supply-chain activities that generate a larger share of the economic returns.

The opportunity may lie beyond building frontier models
The EBRD does not argue that every country should attempt to build its own advanced chip industry or compete directly with the largest AI laboratories. It says the more accessible opportunity for many economies may be downstream: developing AI-enabled applications and adopting AI across existing industries, public services and smaller businesses.
Under a moderate adoption scenario, the report estimates that AI could add between 0.03 and 0.7 percentage points to annual labour-productivity growth across the EBRD regions over the coming decade. The range reflects differences in countries’ ability to adopt the technology, improve skills and modernise infrastructure. It is a scenario estimate, not a guaranteed growth outcome.
For countries with strong agricultural, manufacturing, financial or health sectors, locally adapted applications may offer a more realistic entry point than attempting to develop frontier systems. Examples include tools that help businesses manage inventory, support farmers with crop decisions, improve industrial maintenance or make public services more efficient. The report stresses that the gains depend on practical adoption, trustworthy data and workers who can use the tools effectively.
Defence spending brings another industrial challenge
The report also examines the rise in defence expenditure since Russia’s full-scale invasion of Ukraine in 2022. Higher military demand can create opportunities for domestic manufacturing, engineering, electronics, metals and research. However, the EBRD says most major defence exporters remain outside its regions, while emerging European defence start-ups often face difficulty securing the financing needed to expand.
Drones are identified as a relative bright spot, with companies in Ukraine, Estonia, Poland and Turkey among the faster-growing businesses in the sector. The report also estimates that, on average, a one-euro increase in defence spending in EBRD economies is associated with a subsequent 1.47-euro increase in gross domestic product. That is an observed relationship in the report’s analysis; it should not be read as proof that every additional euro of military spending automatically creates the same amount of economic growth.

Finance, skills and infrastructure are central
The EBRD argues that governments should focus scarce public resources on capabilities that allow local firms to become more competitive. These include research infrastructure, technical and vocational skills, access to finance, stronger links to international supply chains and reliable power. In defence, countries with established industrial strengths may be better placed to expand through retooling, joint procurement and licensing arrangements than by trying to build entire industries from scratch.
The report also points to the importance of entrepreneurship. Start-ups may generate new products and business models, but many struggle to secure later-stage funding and scale beyond their home markets. A deeper venture-capital ecosystem and better connections between founders, investors and established industries could help more promising companies grow.
HeadlinesXtra Analysis
The report’s central message is that the AI race is not only about who develops the most powerful models. It is also about who can provide electricity, computing capacity, skilled labour, capital and the regulatory conditions needed to use AI productively. Countries that cannot compete in advanced chips may still build valuable businesses around local applications, but they need reliable infrastructure and a workforce capable of adapting to technological change.
The same principle applies to defence manufacturing. Increased public spending does not automatically create a competitive industrial base. Without transparent procurement, research partnerships, financing and a clear plan to develop local suppliers, the economic benefits may flow mainly to established foreign producers. Governments will need to balance security needs with social priorities and ensure that public investment builds durable capabilities rather than simply increasing expenditure.
For African economies in particular, the report reinforces the importance of affordable power, broadband, technical education and access to computing resources. The opportunity is real, but it will depend on policy choices and sustained investment rather than the headline growth of the global AI market alone.
Sources and Further Reading
- EBRD Transition Report 2026–27
- EBRD report chapters and executive summary
- EBRD launch release: economies seek new paths to competitiveness
- Reuters report on the EBRD warning, November 17, 2026
- World Bank Group digital strategy for developing countries
- World Bank overview of AI infrastructure readiness