HomeBusinessEBRD Economies Supply Just 0.8% of Global AI Exports, Exposing Competitive Weakness

EBRD Economies Supply Just 0.8% of Global AI Exports, Exposing Competitive Weakness

Featured Summary:

  • EBRD economies account for 8% of global goods exports but supply just 0.8% of AI-related product exports
  • Established semiconductor producers dominate advanced AI manufacturing, limiting opportunities for economies without comparable industrial capabilities
  • Electricity constraints restrict large computing investments, while AI applications offer a less capital-intensive route into technology markets
  • Technology businesses need growth financing and international customers to convert AI development into competitive exports

Economies covered by the European Bank for Reconstruction and Development (EBRD) account for approximately 8% of global merchandise exports but only 0.8% of AI-related product exports.

The figures, published in the bank’s Transition Report 2026–27 on November 17, reveal their limited participation in a growing segment of international technology trade.

Advanced semiconductor manufacturing remains concentrated among established global producers. The EBRD finds that countries entering complex export industries generally build on existing production capabilities.

Afritech Biz Hub Daily Briefings — get the week’s Africa business, tech, and finance signals. Sign up here.

For economies with limited experience in advanced electronics manufacturing, developing competitive AI hardware industries requires industrial expertise and supplier networks that established producers already possess.

Electricity supply also affects where large computing facilities can be developed. Many economies covered by the bank face infrastructure constraints that make such investments difficult.

Technology businesses seeking to expand internationally encounter another obstacle: limited access to the larger financing rounds needed to develop products and enter overseas markets.

The EBRD identifies AI applications and their use across existing industries as more accessible opportunities for many of these economies. Developers can build commercial products using existing AI models without producing advanced chips.

Their prospects for international expansion will depend on whether they can attract investment, secure paying customers and establish demand beyond their domestic markets.

Established Chip Manufacturers Hold Advantages in AI Exports

AI-related goods represented 6.4% of global merchandise trade in 2024, according to the EBRD’s Transition Report 2026–27. Taiwan Semiconductor Manufacturing Company (TSMC) holds approximately 70% of the global semiconductor foundry market, supplying chips designed by major technology companies.

Its manufacturing capacity and established customer relationships give it advantages that newer producers would need substantial investment to challenge.

The EBRD finds that economies entering new export industries generally build on products they already manufacture. Countries with established electronics industries can draw on existing suppliers, technical expertise and production experience to develop more complex products. Those without comparable manufacturing capabilities have fewer opportunities to enter advanced semiconductor supply chains.

Electronics manufacturers in EU member economies covered by the EBRD import almost 60% of their production inputs, compared with approximately 30% in advanced Europe.

The figures indicate differences in the depth of domestic electronics supply chains, although imported components are common in internationally competitive manufacturing. Developing local suppliers capable of producing more sophisticated components would require additional industrial investment.

Advanced semiconductor fabrication requires specialised machinery, highly skilled engineers and manufacturing processes capable of producing chips to demanding specifications.

New facilities must also achieve commercially viable production yields and secure customers willing to place substantial orders. These requirements favour established manufacturers with proven production capabilities, making advanced AI hardware a difficult market for new entrants.

Electricity Constraints Limit Large AI Computing Investments

Global data centres consumed an estimated 485 terawatt-hours of electricity in 2025, approximately 1.5% of worldwide electricity demand, according to figures cited in the EBRD’s Transition Report 2026–27.

The United States and China are projected to account for 78% of additional global data-centre capacity, reflecting the concentration of planned computing expansion in markets with established technology industries.

Electricity shortages restrict opportunities for large data-centre investments across some EBRD economies. Operators require reliable power supplies and grid connections capable of supporting continuous operations.

Where existing networks cannot accommodate additional demand, developers face the expense and delays of securing new electricity infrastructure. High power tariffs can further increase operating costs.

Companies developing AI applications can obtain processing capacity from cloud providers instead of investing in dedicated computing facilities. Existing AI models also allow developers to build commercial software without manufacturing advanced chips or training large models from the beginning.

These arrangements reduce the infrastructure investment required to enter the market, although businesses remain responsible for the cost of using external computing services.

Reliable internet connections and access to computing services remain necessary for developing and operating AI applications. The EBRD identifies digital infrastructure and technical capabilities among the requirements for wider adoption.

Businesses purchasing computing services must also account for usage charges, service availability and the cost of supporting applications as customer demand grows. These expenses influence whether locally developed products can compete commercially beyond their domestic markets.

Foreign Investment Supports AI Startups Facing Expansion Barriers

The EBRD’s Transition Report 2026–27 finds that relatively few startups in its economies grow into large businesses. Venture capital investment remains concentrated in major cities, with companies generally securing less financing than their counterparts in advanced Europe.

Domestic investors are more active in early funding rounds, while businesses raising larger amounts increasingly depend on investors from the United States, United Kingdom and advanced Europe.

Polish robotics company Nomagic raised $44 million in Series B financing in 2025, led by EBRD Venture Capital alongside investors including Khosla Ventures and Almaz Capital.

The company develops AI-powered robotic systems for warehouse operations, serving businesses in logistics, retail and e-commerce. The financing was intended to support further product development and sales expansion.

The EBRD invested $20 million in Ukrainian-founded education platform Preply in 2026 as part of a larger financing round. Preply operates an international language-learning marketplace with more than 100,000 bookable tutors worldwide.

Its expansion includes investment in AI-based personalised learning, although the company’s wider education services mean it should not be classified exclusively as an AI developer.

International financing has helped both businesses pursue commercial expansion, but their activities differ. Nomagic develops AI-powered industrial technology, while Preply operates an established international digital service incorporating AI.

Neither company’s funding announcement, by itself, establishes additional AI-related exports across the EBRD economies. Evidence of overseas sales and sustained revenue growth is needed to assess their contribution to international competitiveness.

The EBRD has also approved investments of up to €25 million in Molten East Fund I and €15 million in Inovo IV, which target technology businesses across parts of Europe and Central Asia.

These funds are not exclusively dedicated to AI development. Their relevance to the region’s competitive position will depend on the companies they finance and whether those businesses secure customers in international markets.

AI Software Exports Could Strengthen EBRD Economies’ Competitiveness

Advanced semiconductor manufacturing remains dominated by established producers, leaving many EBRD economies with limited prospects of competing directly in high-end chip production.

AI software offers another commercial opportunity. Investments in companies such as Nomagic and Preply show that technology businesses in the bank’s regions are attracting capital for product development and international expansion. Their progress in software and technology services, however, is separate from the region’s share of AI-related goods exports.

Under the EBRD’s moderate adoption scenario, AI could increase annual labour productivity growth by 0.03–0.7 percentage points over the coming decade. Businesses would need to adopt the technology effectively to realise those gains.

Improvements in productivity could make domestic industries more competitive, although the projection does not establish how much additional export revenue they might generate.

International sales will provide a clearer measure of whether AI developers can establish lasting positions in overseas markets. Companies securing repeat customers and generating sustained software revenue abroad would demonstrate commercial progress beyond advanced hardware manufacturing.

For EBRD economies, stronger participation in the AI industry will depend on businesses turning technological capabilities into products that can compete internationally.

Ajibola Olarinoye
Ajibola Olarinoyehttps://afritechbizhub.com/
Ajibola Olarinoye is a Mass Communication graduate with over ten years of experience in content writing and journalism. He specialises in business, technology, lifestyle, politics, and political economy, producing clear, engaging, and informative articles that simplify complex issues, highlight emerging trends, and provide readers with valuable insights.
RELATED ARTICLES

Most Popular

Recent Comments