AI Spending Fuels Historic Computer Investment Surge, But Growth Shows Signs of Cooling

US business investment in computers surged 115% in 18 months, driven by AI infrastructure spending that now accounts for one-third of GDP growth, though momentum cooled significantly in Q2 2026.

Illustrative image only — not an actual photo of the events or people described in this article.

The chart reveals two distinct eras. From early 2021 through late 2023, computer investment bounced between $143 billion and $165 billion per quarter, showing modest fluctuations but no clear direction. Then everything changed in Q4 2023, when spending hit $155.1 billion and began a relentless climb. By Q1 2024, it reached $165.5 billion. Three quarters later, in Q4 2024, it stood at $186.1 billion. But the real acceleration came next: Q1 2025 jumped to $222.8 billion, Q2 2025 surged to $253.8 billion, Q3 2025 climbed to $280.9 billion, and Q4 2025 reached $325.4 billion. The first quarter of 2026 saw investment rocket to $384.2 billion, an 18% increase in just three months. Then came Q2 2026 at $400.6 billion—still growing, but at a much gentler 4.3% pace compared to the previous quarter's explosive gains.

This investment boom isn't happening in a vacuum. Tech giants collectively poured an estimated $320 billion into AI infrastructure in 2025 alone, with plans to spend $400 billion in 2026. The ING analysis shows that when you account for computing, peripheral equipment, software investment, and data center construction—then subtract the imported technology that doesn't contribute to US GDP—AI-related tech investment accounted for roughly one-third of year-over-year GDP growth in Q2 2026. Amazon committed to approximately $200 billion in capital expenditures for 2026, representing a 60% increase from the previous year. This spending spree has created what analysts call an "investment supercycle," with global capacity projected to require $3 trillion in total capital by decade's end.

But the flattening growth rate from Q1 to Q2 2026 raises important questions about sustainability. While investment hit a new peak, the quarter-over-quarter increase slowed dramatically—from 18% to just 4.3%. This deceleration coincided with broader economic headwinds: US GDP growth slowed to 1.5% in Q2 2026, down from 2.1% in Q1 and missing the 2.1% consensus forecast. The slowdown wasn't due to weak domestic demand—consumer spending and business investment actually strengthened. Instead, a sharp 11.5% increase in imports, driven partly by heavy purchases of semiconductors and other goods feeding the AI buildout, cut 1.5 percentage points from GDP. Since chips account for roughly 60% of total AI data center investment, according to McKinsey, much of that computer equipment spending flows right back overseas to semiconductor manufacturers. The result is a widening trade deficit that creates a mathematical drag on GDP even as companies spend furiously on AI infrastructure.

The sharp climb from late 2024 through early 2026 captured the initial frenzy of AI infrastructure buildout—the moment when every major tech company realized it needed massive computing capacity immediately or risk falling behind. The cooling pace in Q2 2026 doesn't signal retreat; it suggests the market is catching its breath after an unsustainable sprint. Companies are still investing at record absolute levels, but the rate of acceleration has moderated as supply chains strain, power constraints emerge, and firms start questioning whether current spending can generate returns before the next technological shift arrives. What happens next depends on whether AI applications justify the infrastructure already built—or whether this becomes a cautionary tale about building too much, too fast.