East Asia’s manufacturing growth slows as AI alone can’t revive it

EAST ASIA (INCLUDING MALAYSIA, VIETNAM, THAILAND, PHILIPPINES), 11 OCTOBER 2026 —

East Asia’s AI hardware boom stalls as regional economies struggle to capture value, World Bank warns

Malaysia and Southeast Asia’s AI-driven export surge risks leaving local firms and workers behind as global rules favour tech giants, according to a World Bank report.

The rapid expansion of AI-related manufacturing in East Asia—driven by a US$1.4 trillion surge in exports over the past year—has yet to translate into broad-based economic benefits for the region’s developing nations, the World Bank’s latest economic update reveals. While countries like Malaysia, Vietnam, and Thailand have become critical suppliers of AI hardware, the value captured remains concentrated in foreign-owned design, software, and intellectual property, leaving local firms and workers with limited gains.

The report underscores a critical paradox: as AI hardware exports grow, the traditional manufacturing "escalator" that once lifted economies from agriculture to higher-value industries has stalled, and the rules governing AI’s economic future are being written without meaningful regional input.

The World Bank’s October East Asia and Pacific Economic Update, titled "Riding the AI Wave", highlights how AI-related exports—including computers, servers, and routers—have surged, particularly in Vietnam and Malaysia, where shipments have more than doubled since 2020–22. US imports of AI-enabling goods rose nearly 60% in early 2026, while other imports declined by nearly a quarter, reflecting a global shift toward AI infrastructure.

Yet the report warns that this growth is uneven, with China accounting for US$785 billion in AI-related exports—mostly intermediates shipped to ASEAN nations for final assembly—while the intellectual property and software driving these products remain largely foreign-controlled.

For Malaysia, the economic calculus is stark. The report notes that each data centre job requires RM4.6 million in investment, a figure far exceeding the RM1 million needed for electronics or automotive manufacturing roles. This disparity underscores a broader challenge: AI hardware exports, while boosting GDP, have not translated into the job creation or income growth that traditional manufacturing once delivered.

The World Bank’s own analysis suggests that countries reaching peak industrial employment shares in 2015 and 2019 did so at per capita incomes ranging from US$900 to US$5,900—far below high-income thresholds. Meanwhile, the report points to services—such as software development, business process outsourcing, and data work—as the next frontier for growth, yet these sectors are largely absent from current AI trade data.

The Philippines and Vietnam are leading the charge in services, with Vietnam’s World Trade Organization (WTO) accession accelerating productivity gains in both manufacturing and service sectors. However, the report cautions that the region’s AI-driven growth remains vulnerable to external shocks.

The US$1 trillion investment cycle by major hyperscalers over 2025–26 could reverse if demand lags, while the rules governing AI’s economic landscape—including cross-border data flows, digital services, and export controls—are being shaped unilaterally by the US and other powers. Developing nations, the report argues, have little say in these decisions, which risk entrenching a system where the terms of trade favour a handful of tech giants.

Mia Mikic, a trade economist and member of the Friends of Multilateralism Group, frames the issue as a battle for influence. "The economic terms of the AI economy are being set somewhere else, and mostly outside multilateral forums," she said.

"The region’s AI export boom rests on unilateral decisions of importing markets—decisions that can be withdrawn just as easily." The report calls for emerging economies to shift their competitive strategies, emphasizing openness to services, technology, and ideas rather than relying solely on cheap labour or economies of scale.

It also urges regional nations to advocate more aggressively for a fairer multilateral rulebook, lest they remain spectators in an economy increasingly dominated by a few powerful actors.

For Malaysia and its ASEAN peers, the implications are clear: the AI hardware boom is not a silver bullet for inclusive growth. Without deeper integration into high-value services, stronger domestic innovation ecosystems, and a greater voice in shaping global AI standards, the region risks becoming a peripheral player in an industry that promises to reshape the global economy.

The World Bank’s findings suggest that the next phase of economic development will demand more than just assembling the world’s AI infrastructure—it will require redefining how value is created, shared, and controlled.

Related: World Bank · Mia Mikic