The Real Artificial Intelligence Boom Hasn't Yet Begun

The biggest part of the artificial-intelligence revolution is still ahead of us.

Technology companies are spending staggering sums on chips, data centers, electricity and increasingly powerful AI models. Investors are understandably asking whether too much money has been committed too quickly. Some investments will disappoint. Some data centers may prove unnecessary, and some AI valuations may eventually look absurd.

But those questions obscure something more important. The technology industry has accomplished something extraordinary: It is making knowledge, analysis and problem-solving capability available to anyone, instantly.

The internet put information at our fingertips. Artificial intelligence goes a step further. It helps us understand, analyze, and act on that information.

Now comes the next stage of the revolution: figuring out what millions of people and businesses will do with it.

That is how transformative technologies develop. Electricity did not change the world simply because utilities learned how to generate and distribute power. The larger transformation occurred when manufacturers, inventors and entrepreneurs discovered what that power could do.

Factories were redesigned around electric motors. Entrepreneurs created refrigerators, washing machines, radios, televisions and air conditioners. Electricity transformed communications, medicine, transportation, entertainment and almost every American home.

The electric company did not invent all those products. It supplied the power that made them possible.

Artificial intelligence is entering the same stage.

Chipmakers provide the computing machinery. Data centers provide enormous processing capacity. Companies such as OpenAI, Google and Anthropic have built platforms capable of delivering increasingly sophisticated knowledge, analysis and assistance to almost anyone with an internet connection.

Electric utilities made power available on demand. AI companies are beginning to make something equally consequential available on demand: the ability to retrieve knowledge, analyze information, solve problems and assist with decisions.

For simplicity, call it intelligence on demand.

The next great opportunity is figuring out what to do with it.

I use artificial intelligence every day and have spent considerable time studying and experimenting with it. Yet I continually discover capabilities I did not know existed. I still find myself asking how I can use AI more effectively in my business, my home and the organizations with which I am involved.

That experience tells me that one of the greatest obstacles to AI adoption is no longer what the technology can do. It is that most people and organizations still do not understand everything they can do with it.

That gap will create an enormous industry.

We tend to think about the AI economy in terms of chips, data centers and models. But another layer is now emerging: application and implementation.

This is where thousands of entrepreneurs and businesses will enter the picture. They will not need to create new foundation models or build enormous data centers. Instead, they will take the capabilities already being produced and apply them to real problems in particular industries, professions, schools and homes.

For businesses, that means going far beyond teaching employees how to write better prompts.

Someone needs to understand how a company actually operates. What are employees doing all day? What information are they repeatedly searching for? What reports take hours to prepare? Where do mistakes occur? Which processes are repetitive? What valuable knowledge is buried in old files or in the heads of longtime employees? Which decisions can be made faster or better if information is easier to retrieve and analyze?

Then comes the important question: What can AI do about it?

Consider real estate. Artificial intelligence can read leases, identify rent escalations and expiration dates, analyze operating expenses, prepare reports, track maintenance problems, assist with tenant communications, evaluate acquisitions and search decades of company records.

But handing every employee a subscription to an AI chatbot does not make any of that happen.

Someone has to connect the intelligence to the business.

Multiply that challenge across millions of companies and a major new industry emerges: AI consultants, workflow designers, systems integrators, trainers and specialized companies that understand both artificial intelligence and the industries they serve.

Education will present an even greater opportunity.

Much of today's discussion focuses on preventing students from using AI to cheat. That concern is legitimate, but schools should spend at least as much time teaching students how to use AI to learn.

Imagine a student struggling with algebra who tells an AI tutor: Do not give me the answer. Determine what I do not understand. Explain it another way. Give me another example. Ask me questions. Do not move on until I understand the concept.

That is not replacing education. It is providing something generations of parents would have paid dearly to give their children: an infinitely patient personal tutor available at any hour.

Students will need to learn how to use AI that way. Teachers will need to learn how to teach with it. Universities will have to rethink curricula, assignments and the skills graduates will require when nearly every profession has access to inexpensive artificial intelligence.

The same transformation will reach our homes.

Artificial intelligence will be embedded in cars, appliances, financial services, healthcare, travel and entertainment. Eventually consumers will care as little about which AI model operates behind a product as they now care which power plant supplies the electricity when they flip on a light.

AI will manage calendars, organize records, research purchases, plan vacations, monitor household systems, interact with automobiles and perform countless administrative chores. Families will spend routinely on AI-enabled services just as they now spend on broadband and smartphones, sometimes through subscriptions and sometimes because the intelligence is embedded in the products and services they buy.

This is why the next generation of important AI companies will look very different from today's.

Some will specialize in medicine, law, accounting, manufacturing or real estate. Others will specialize in schools and universities. Large consulting firms will participate, but so will thousands of smaller companies that understand particular industries better than the largest technology companies ever could.

We will also see personal AI advisers who study how individuals and families live and work and show them how artificial intelligence can save time, improve decisions and eliminate unnecessary work.

This transformation also complicates predictions that AI will simply eliminate employment. It will certainly replace tasks and some occupations. But transformative technologies create industries as well as disrupt them. Electrification did not merely replace steam-powered machinery. It enabled entirely new businesses and occupations that could not have existed before widespread electric power.

Artificial intelligence will do the same.

That is why today's debate over an AI bubble risks confusing the price of today's investments with the ultimate importance of the technology.

The dot-com era offers a useful lesson. Investors dramatically overvalued many internet companies. They did not overestimate the internet. If anything, they underestimated how completely it would transform commerce, communications, entertainment and everyday life.

We may look back and conclude that some companies built too many data centers too quickly or that investors paid too much for certain AI businesses. At the same time, we will discover that we badly underestimated what happens when inexpensive knowledge, analysis and problem-solving capability become available to nearly everyone.

Electricity gave the industrial economy power on demand. Artificial intelligence will give the next economy intelligence on demand.

The companies building that intelligence will create enormous value. But the next great wave of opportunity will belong to the entrepreneurs, businesses, teachers and inventors who discover what to do with it.