Investment in artificial intelligence is giving the United States economy a boost, but economists see little proof that the country has entered a new period of faster long-term growth. Spending on data centres, computer equipment and AI infrastructure has risen sharply as businesses race to build systems that can handle more advanced software. That spending supports construction, technology suppliers and related industries, yet the wider effect on productivity remains uncertain.
President Donald Trump has promised a stronger economic era through tariffs, tighter immigration rules, tax cuts and deregulation. His administration argues that these policies will encourage companies to invest at home, create jobs and raise output. However, recent growth has remained broadly in line with the pattern seen in previous years. Economists estimate that the underlying growth rate of the US economy is around 2%, reflecting factors such as an ageing population and slower immigration.
AI investment has become one of the clearest sources of new spending. Technology firms and other businesses are building data centres and buying advanced chips to train and operate AI systems. These projects require large amounts of construction work, electricity, cooling equipment and network capacity. They also create demand for specialist workers and services. As a result, AI spending can lift economic activity even before companies show that the technology has greatly improved their productivity.
The key question is whether the spending will lead to lasting gains. If AI tools allow workers to complete tasks faster, reduce errors and develop new products, businesses could produce more with the same amount of time and resources. Such gains could support higher wages and stronger economic growth. However, economists caution that early productivity figures can be volatile and may reflect factors other than AI. It is too soon to conclude that the technology has changed the long-term growth path of the economy.
The effects of tariffs have also been mixed. The administration has argued that higher duties on imported goods will encourage manufacturers to bring production back to the United States. Yet imports have continued to rise, partly because technology companies need equipment and components for AI data centres. Tariffs can increase costs for businesses that depend on foreign parts, and companies may pass some of those costs to customers. That can make it harder to judge whether the policies are delivering the promised industrial revival.
The expansion of data centres has also created local concerns. These facilities can use large amounts of electricity and water, and communities have questioned whether the benefits justify the extra pressure on power systems. Higher electricity demand can affect household bills if supply does not grow quickly enough. Some local authorities have faced resistance to new projects, making it harder for companies to expand as quickly as they planned.
Tax incentives have helped support investment in data centres and AI infrastructure. The government hopes that private spending will eventually lead to wider productivity gains, stronger output and better public finances. But the Congressional Budget Office expects potential growth to remain near 1.8%, suggesting that a lasting transformation has not yet been established.
For businesses, the current period offers both opportunity and risk. Companies that invest early may gain an advantage if AI tools prove useful at scale. Others may spend heavily without seeing enough return. Investors will also need to distinguish between firms with clear business uses for AI and those relying mainly on market excitement.
AI is already contributing to US economic activity, but the larger promise remains unproven. The next few years will show whether investment in computing systems becomes a lasting source of productivity or mainly a temporary spending boom. For now, the technology is helping growth, while economists remain cautious about claims that it has permanently changed the economy’s potential.
