# AI Investment Paradox: $410 Billion With No Impact on GDP
Investments in artificial intelligence reached $410 billion in 2025, but their contribution to US GDP has been virtually zero. An NBER survey of 6,000 company executives in the US, UK, Germany, and Australia revealed that 80% of firms saw no impact from AI on productivity or employment over the past three years. Yet 69% of companies are already using AI, and managers spend an average of 1.5 hours per week on it.
Economists attribute this to a lag in monetization: AI speeds up individual tasks, but financial results trail behind. Goldman Sachs chief economist Jan Hatzius pointed out that the money has gone toward importing chips from Taiwan and Korea, boosting their economies instead of America's.
NBER Survey: Data and Findings
The NBER study found that subjective perceptions of AI benefits outpace objective metrics. Executives feel processes speeding up, but it's not showing in the numbers. Key figures:
- 69% of companies use AI.
- 80% saw no changes in productivity or hiring.
- Managers: 1.5 hours/week on AI.
- Forecast: +1.4% productivity and +0.8% output in 3 years.
Dario Perkins of TS Lombard told the Financial Times: there's no evidence of AI impacting productivity or employment in the US. Strong labor figures are tied to the business cycle, not automation.
Comparison with Historical Parallels
NBER economists draw parallels to the computer revolution of the 1970s–1980s. Massive PC investments only boosted productivity after 20+ years. Similarly, current AI spending ($660 billion forecast for 2026 from Bridgewater) may pay off with a delay.
Investors remain optimistic despite the paradox. Executives expect returns in the coming years, fueling continued capital inflows.
Key Takeaways
- Effect Lag: AI accelerates tasks but doesn't immediately boost GDP due to supply chains (chip imports).
- Survey Data: 80% of firms see no growth metrics, but 69% are using AI.
- Forecasts: +1.4% productivity in 3 years per CEO estimates.
- Historical Parallel: 20 years for computers to show effects.
- Investments Growing: $410 billion in 2025 → $660 billion in 2026.
Implications for Business and Markets
For companies, the paradox means focusing on integrating AI into core processes. Current spending is often scattered, without a systematic approach to measuring ROI. Economists recommend tracking not just costs, but intermediate metrics: time on tasks, output quality, scalability.
At the macro level, AI's impact may show up in redistribution: Taiwan and Korea benefit from hardware exports, while the US invests in software and data. This exacerbates global imbalances in supply chains.
In the long term, the optimism is justified: like with PCs, combining hardware, software, and workforce retraining will deliver a boost. But expecting quick GDP growth is premature.
— Editorial Team
No comments yet.