Silicon Valley Discovers That Computers Capable of Replacing Human Intelligence Have Not Yet Learned How to Replace Human Revenue
NEW YORK, October 17, 2026. Wall Street has experienced a disturbing encounter with arithmetic after investors began questioning whether the artificial intelligence industry might eventually need to earn enough money to pay for its extraordinary expenses.
Nasdaq Falls 2.1% as AI Spending Worries Return
American technology stocks suffered substantial losses Friday, with the Nasdaq Composite falling 2.1% amid renewed concern about the financial sustainability of enormous AI investments.
According to Associated Press report on the October 16 technology stock market decline, major technology and semiconductor companies led the market downturn.
For several years, investors have enthusiastically financed systems capable of drafting emails, answering questions, producing software and generating detailed business strategies.
The machines have performed admirably.
The companies operating them, however, must still pay for electricity, equipment, buildings and thousands of highly compensated engineers. The U.S. Energy Information Administration publishes what that electricity costs, and nobody in Silicon Valley has been reading it with a smile.
Wall Street considers these expenses an unexpected complication.
One technology investment analyst explained the industry’s underlying financial philosophy.
“We assumed that if a computer became intelligent enough, it would eventually figure out how to make money. We didn’t expect investors to ask whether the companies had already done that.”
Revolutionary Technology Meets Quarterly Earnings
Silicon Valley executives insist artificial intelligence will transform the global economy.
They have offered convincing demonstrations involving medical research, software development and advanced scientific modeling.
Yet investors have discovered that transforming the global economy is not necessarily identical to generating a quarterly profit.
A company can possess revolutionary technology, millions of enthusiastic users and a computer system capable of composing a symphony.
If the electricity bill exceeds revenue, the accountant remains unimpressed.
The situation has produced an unusual paradox.
Artificial intelligence is marketed as a technology that will make businesses dramatically more efficient.
Developing that efficiency apparently requires spending unprecedented sums of money.
One Wall Street economist compared the arrangement to building a machine that produces free sandwiches.
“The machine is extraordinary. It makes sandwiches in two seconds. It also costs fourteen billion dollars and requires a nuclear reactor to operate.”
Investors Ask Whether the Next Generation of AI Will Be Cheaper
Technology executives have responded by promising that the next generation of systems will be even more powerful.
Investors have responded by asking whether the next generation might also be cheaper.
The question has introduced considerable tension into corporate earnings presentations.
For years, investors rewarded companies simply for mentioning artificial intelligence repeatedly.
A traditional manufacturing firm could announce a modest AI initiative and immediately attract speculative enthusiasm.
A restaurant chain could describe its computerized reservation system as an artificial intelligence transformation.
A plumbing company could rename its appointment calendar a predictive infrastructure platform.
Everyone benefited.
Except perhaps the accountants.
Now investors are demanding evidence of commercial returns, sustainable demand and reasonable infrastructure costs. The Securities and Exchange Commission’s investor guidance has recommended asking exactly those questions for roughly ninety years.
Silicon Valley considers this an abrupt change in the rules.
One venture capitalist complained that the public has become excessively obsessed with profitability.
“We’re building the future of civilization. It’s unreasonable to interrupt every presentation with questions about operating margins.”
Technology Bubbles, Profitability and the Dot-Com Lesson
The controversy highlights an enduring difficulty in financial markets.
Technological innovation can be genuinely valuable without every company associated with it becoming a successful investment.
The internet transformed civilization, but that did not make every internet business profitable. Anyone who owned shares during the dot-com bubble already knows how that chapter ends.
Artificial intelligence might follow a similar trajectory, producing extraordinary benefits alongside extremely expensive disappointments.
Meanwhile, electricity providers appear delighted.
Unlike venture capitalists, they have discovered a business model in which customers must actually pay for the product.
Wall Street now faces its most uncomfortable technological revelation.
Artificial intelligence may eventually replace analysts, consultants and several categories of office workers.
But somebody will still need to calculate whether the whole operation makes financial sense.
For the moment, that responsibility remains with the one person Silicon Valley cannot afford to replace.
The accountant.
Five Humorous Observations
- A machine that writes 10,000 business plans and not one of them says “pay the electric bill.” It can write a symphony, but it can’t read a utility envelope.
- Fourteen billion dollars and a nuclear reactor for a sandwich. My cousin makes a sandwich for free and doesn’t need a reactor. Mostly just a bad mood.
- A plumbing company renames its calendar a predictive infrastructure platform. It’s the same guy. He still shows up Thursday, sometime between eight and four.
- “It’s unreasonable to interrupt every presentation with questions about operating margins.” That’s what every guy says right before he asks you to wire him money.
- The accountant is the one job they can’t replace. Meanwhile he’s been sitting at the same desk since the nineties saying “that doesn’t add up,” and nobody listened until now.