Bank of America Global Research, in the first installment of a series titled "AI Matters," projects that investment in artificial intelligence will remain a cornerstone of global economic expansion through 2026. The report attributes roughly 0.4 percentage points of US GDP growth this year to AI-related capital expenditure, underscoring the technology's current role as a macroeconomic accelerator.
That contribution, the report states, is expected to wane in 2027 even as the spending plans of large cloud and AI infrastructure firms - described in the report as "hyperscalers" - introduce significant upside risks to the near-term bullish case. The analysis also points out that this investment cycle is not confined to the domestic market; suppliers in economies such as Taiwan, Mexico, and Korea are emerging as key beneficiaries as capital flows shift toward the hardware and infrastructure needed to support AI deployment.
Competing models: innovation versus scale
The report frames the US and China as engaged in an intense contest for AI primacy, but following distinct strategies. Washington is characterized as the leader in frontier model research and development, driven by private-sector dynamism and deep research capabilities. In contrast, Beijing is depicted as prioritizing state-led scaling and close control over manufacturing, with advantages that include lower energy costs and centralized access to critical minerals required for hardware production.
These divergent approaches are producing a visible ripple effect across global supply chains. The surge in AI spending from both countries is creating a tailwind for major exporters of semiconductors and other infrastructure components. The report notes analysts have maintained an 8% GDP forecast for Taiwan for 2026, with the AI sector's expansion cited as a chief reason for that optimism.
Despite geopolitical tensions in the region, demand for high-end chips and the physical infrastructure to support large-scale AI workloads remains robust. Mexico and Korea are highlighted as beneficiaries that are realizing structural gains as they integrate more deeply into the global AI hardware supply chain. Collectively, these trends indicate the economic effects of AI are moving beyond a Silicon Valley-centric story toward a more geographically diversified growth engine.
From capex to productivity: the next phase
While the initial phase of AI's economic influence is anchored in infrastructure and hardware investments, the report stresses that the subsequent phase will hinge on deployment across domestic workforces. BofA analysts are closely tracking whether AI will produce only incremental productivity improvements or a more fundamental step-change in how labor is organized and valued.
The report warns that some sectors confront displacement risks as automation and AI tools are introduced, but it emphasizes a broader "skills challenge" that could determine national competitiveness into the late 2020s. The ability of firms and workers to adopt AI tools effectively will shape how much of the initial capital outlay translates into lasting productivity gains.
Moreover, the investment cycle appears not to have peaked. Continued capital expenditure targeted at data centers and specialized silicon will place a premium on the capacity of global suppliers to meet rising demand. The report underscores that investors are likely to concentrate on "top-tier" economies able to sustain their roles as essential nodes in the evolving technological architecture, since supplier readiness will be a defining element of international growth benchmarks.
Implications for markets and policy
By combining a near-term GDP contribution from AI capex with the prospect of longer-term productivity shifts, the report frames AI as both an immediate market stimulant and a strategic battleground. The more capital flows into infrastructure and silicon, the more important the global distribution of suppliers and manufacturing capacity becomes for macroeconomic outcomes.
The report leaves open whether current investment patterns will culminate in widespread labor-market transformation or a more gradual augmentation of existing roles, but it clearly identifies the skills pipeline and supplier capacity as central uncertainties for the decade ahead.