Railroads and the internet may offer insight into the path ahead with AI

Although much is still unknown about the long-term implications of the mass investment in artificial intelligence (AI) that’s underway, its impact already is undeniable.

AI is driving one of the largest capital investment cycles in decades, fueling spending on data centers, semiconductors and infrastructure and raising questions about productivity, jobs and long-term returns. In this environment, focusing too narrowly on short-term risks, like whether companies may be overinvesting, can obscure the broader transformation underway, BOK Financial® experts said.

The bigger picture is that AI is driving a broad capital expenditure cycle that extends beyond technology into manufacturing, infrastructure and energy—positioning AI and automation together as a key engine of economic growth over the next 10 years, said Matt Stephani, president of Cavanal Hill Investment Management, Inc., a subsidiary of BOK Financial Corporation.

“These technologies ultimately are deflationary,” he explained. “You can have lower labor costs per unit of production as you automate and use AI. I think there’s this bet that AI is going to lead to a big productivity gain—and I share that view.”

However, as in any transformation, there might be some pain along the way.

As BOK Financial Chief Investment Officer Brian Henderson explained, “Because of the buildout of AI infrastructure and the demand for semiconductors and electricity, it’s actually boosting inflation in the short term.”

Over the longer term, however, those same investments are expected to lower costs by improving efficiency, which has major implications for individuals, businesses, financial markets and the overall U.S. economy. “It has the potential to increase productivity, allow the economy to grow faster and make jobs more efficient,” he said.

Similarly, as companies implement AI and automation, there is the understandable fear that people are going to lose their jobs if their roles become obsolete, said BOK Financial Chief Investment Strategist Steve Wyett.

Historically, similar technological shifts—from railroads to the internet—have displaced workers in the short term while creating new opportunities over time, he explained, noting that AI appears likely to follow a similar path. Consequently, workers may need to learn different skills or adapt as certain roles evolve or disappear. In other words: “There still are going to be jobs out there; they just may be different.”

AI, automation already changing the face of work

Already, companies of all sizes are dedicating resources to AI, with the hope that this investment will give them an edge not just now but also in the future—and many companies are already seeing this impact.

As it is now, only 37% of the companies included in Deloitte’s 2026 report, The State of AI in the Enterprise, say that their organization’s use of AI is surface level. Instead, most enterprises are using AI on a deeper level, with 30% reporting that it’s “redesigning key processes” and 34% saying that it’s “deeply transforming” their business. Meanwhile, employees, too, are experiencing the effects of enterprise AI adoption, as workers’ access to AI grew by 50% in 2025, the Deloitte report found.

As with any adoption of new technology, there is the potential for overinvestment and other mistakes. For instance, a company may exhaust their annual AI budget in just a few months, and not all companies will survive this transformation, but this is all to be expected, experts said.

As the labor pool shrinks, AI may be the answer

Looking forward, one of the biggest beneficiaries of AI and automation may be the U.S. economy itself, experts agreed. The potential boost to productivity from these technologies may be crucial as the U.S. faces slowing labor force growth due to demographics and immigration trends. By 2030, the number of Americans aged 65 and older is expected to surpass those aged 0-17, reversing a generations-long trend, according to the U.S. Census Bureau’s national population projections.

This means that productivity gains are essential to sustaining U.S. economic expansion. As Wyett said, “We may need every bit of AI’s productivity because our labor force is going to be shrinking over time.”

Stephani reinforced this view, noting, “As you automate and use AI, you lower the cost—and that productivity is what supports growth over time.”

Change is happening both fast and slow

On one hand, changes from AI are already around us. As Henderson said, “AI has become a huge factor in the financial markets, particularly the performance of the stock market, and it has been a huge contributor to overall economic growth.”

On the other hand, it will take time for AI’s impact to fully unfold, he noted. The current phase of AI development and implementation—marked by heavy investment, rapid innovation and some degree of uncertainty—is just the beginning.

History can again provide some insight into what to expect. As Wyett said, transformative technologies such as railroads, the internet and now AI often involve “big losses” early on but ultimately reshape how the economy operates.”

“Think of the internet coming in the mid-90s,” he continued. “There was this huge bubble with big losses. However, as we sit here today, the internet is ubiquitous. We can’t operate without it.”

Read more of our 2026 Midyear Market Update.