What is the effect of AI capital expenditures on the US GDP growth trajectory

Barclays analysis indicates that AI capital expenditures by major U.S. tech firms, particularly hyperscalers, contributed approximately 1 percentage point to U.S. GDP growth in the first half of 2025, primarily driven by spending on computer equipment and software. However, this economic boost is projected to be short-lived, with the GDP contribution from AI-sensitive investments expected to peak this year and then decline significantly to 0.2 percentage points by 2027 due to a deceleration in hyperscaler capital expenditure growth. This suggests that while AI-driven capital outlays provide a clear short-term economic lift, their potential for sustained long-term GDP growth or productivity gains is limited.

Analysis

Barclays analysis indicates that capital expenditures on artificial intelligence by major U.S. technology firms, particularly hyperscalers like Amazon, Microsoft, and Google, contributed approximately 1 percentage point to U.S. GDP growth in the first half of 2025. This boost was primarily driven by spending on computer equipment and software, with data center investment having peaked in early 2023. AI-sensitive investments accounted for nearly all of the 8.1% SAAR rise in aggregate business fixed investment during H1 2025.

Despite this significant short-term impact, the growth momentum from AI-sensitive investments is projected to ease in the medium term. Hyperscaler capital outlays are forecast to grow 30% from $370 billion in 2025 to $510 billion by 2027, a notable deceleration from the 60-70% annualized growth rates observed between 2023 and 2025. Consequently, Barclays projects the GDP contribution from AI-sensitive investments to peak at 1.0 percentage point this year, then decline to 0.55 percentage point in 2026 and 0.2 percentage point in 2027.

The analysis suggests that AI capital expenditures have limited potential to sustain long-term GDP growth or productivity gains. While significant in scale, hyperscaler spending represents a modest share of total U.S. business fixed investment, and achieving structural productivity increases would require a 20% rise in BFI, a level not recently observed. Even planned complementary infrastructure spending by utilities, totaling $1.1 trillion from 2025-2029, is estimated to have a small, diminishing GDP contribution.

Key Decisions for Investors

• Investors should re-evaluate long-term growth expectations for companies heavily reliant on AI-driven capital expenditure, given the projected deceleration in hyperscaler spending and its diminishing GDP contribution post-2025.
• Consider shifting focus from direct AI infrastructure spending beneficiaries to companies demonstrating clear, sustainable productivity gains from AI adoption, as the macro impact of capex alone is expected to wane.
• Closely monitor broader business fixed investment (BFI) trends beyond AI-specific outlays, as the article highlights AI's modest share of total BFI and the high bar for sustained structural productivity improvements.