The Revolution May Be Delayed: AI, the Dot-Com Bubble, and the Missing Foundation

History rarely repeats itself exactly, but its rhymes are unmistakable. The current mania surrounding Artificial Intelligence—a frenzy of impossible valuations and world-changing promises—is an echo of the dot-com bubble. In both cases, the technology was, and is, genuinely revolutionary. But in both cases, the hype arrived before the foundation needed to support it.

The playbook is identical: a transformative idea is mistaken for a transformative business. We forget that technologies like the radio, the television, and the internet itself were fascinating novelties long before they were economic engines. They only reshaped society after the boring, expensive, and unglamorous work of building infrastructure was complete, and someone finally cracked the code of monetization.

We feel the same mistakes are being made today.

The late 1990s promised a “new economy.” We envisioned a world of frictionless e-commerce, instant global communication, and a library of all human knowledge in every home. The companies that embodied this promise, like Pets.com or Webvan, weren’t necessarily bad ideas—they were just impossible ideas for the era.

• The Missing Infrastructure: The “information superhighway” was a dirt road. The vast majority of users were on screeching 56k dial-up modems. Secure online payments were novel and mistrusted. Logistics networks for one-day delivery didn’t exist. The promise of “streaming video” was a 10-second, postage-stamp-sized clip that buffered for five minutes.

• The Missing Monetization: With no reliable payment systems and a tiny user base, the only viable business model was “get eyeballs.” Companies burned billions in venture capital on Super Bowl ads, hoping to figure out how to make money later. They never got the chance.

The bubble burst, not because the internet was a fad, but because its “transformative potential” was trapped behind a massive infrastructure bottleneck. The real revolution only arrived a decade later, built on the fiber-optic cable laid in the bubble’s ruins and powered by broadband, 4G, and the cloud.

Today, AI promises to be our collective super brain—a productivity engine that will cure diseases, write code, and discover new materials. We see its potential in dazzling demos from ChatGPT and image generators. But just like in 1999, we are mistaking the demo for the product.

• The Missing Infrastructure: The “picks and shovels” narrative of today, centered on selling GPUs, masks a terrifying reality: we lack the power to run this revolution at scale. The energy consumption of today’s models is staggering, and training next-generation models may require the output of entire nuclear power plants. We are building supercars, but we haven’t built a national highway system, and we don’t have nearly enough gas stations. The cost of compute is so high that running an “AI-powered” search query costs 10 to 100 times more than a standard one—a cost no one has a plan to pay for.

• The Missing Monetization: The current business models are just as shaky as “get eyeballs.” The dominant model is “burn billions in R&D and charge a $20/month subscription.” Companies worldwide are being sold on “integrating AI,” but most are just experimenting, with no clear return on investment. The high compute costs mean that profitability remains a distant dream for most AI applications.

The dot-com crash was a painful but necessary correction that separated the idea of the internet from the business of the internet. Are we hurtling toward a similar “AI winter”? The technology is real. The transformation is coming. But it may only be built after this bubble bursts, after we solve the monumental energy and infrastructure crisis, and after sustainable business models emerge from the rubble.

In the long-term we’re optimistic about the sector, but it’s impossible to know which of the players will still be viable. As always work closely with your Certified Financial Planner® professional to determine what level of exposure is appropriate for you. 

Stephen Kyne, CFP® is a Partner at Sterling Manor Financial, LLC in Saratoga Springs.

Sterling Manor Financial, LLC is an SEC Registered Investment Advisor and does not provide tax or legal advice, nor is it a third-party administrator. Consult your attorney or accountant prior to implementing any tax or legal strategies. Article contains forward- looking statements and opinions which are subject to change.