1. Why Power Became Institutional

For most of modern economic history, power became concentrated inside institutions because institutions solved a basic problem: scale. They pooled capital, expertise, labor, infrastructure, and distribution. A company could accomplish things that an individual simply could not. As a result, organizational size became closely associated with productive capability—and productive capability became power.

2. AI Starts Deflating the Structure

AI begins to weaken that relationship. It does not merely automate individual tasks; it can reduce the amount of human and organizational machinery required to produce an outcome. Research, analysis, writing, coding, design, customer support, marketing, and administration can increasingly be performed with far fewer people. This is what I mean by structural deflation: the shrinking of the human and organizational structure needed to create economic output.

3. The One-Person Company

The clearest expression of this change is the one-person company. An individual can combine AI with cloud infrastructure and digital platforms to perform functions that once required departments. AI can assist with research, software development, design, marketing, documentation, analysis, and operations. The human still provides judgment, direction, accountability, and the decision about what should be built—but the amount of organizational machinery surrounding that person can become dramatically smaller.

4. When One Can Compete With 100

Large companies will continue to possess advantages that AI cannot instantly reproduce: capital, brands, relationships, physical infrastructure, regulatory capacity, and established distribution. But in knowledge-intensive and digital businesses, headcount becomes a less decisive source of advantage. If a small team can use AI to perform work that previously required dozens or hundreds of employees, the minimum efficient scale of the business falls.

That is a deeper disruption than simple job automation. AI changes the economics of organizational size. Large organizations can automate, but smaller organizations become viable in areas that previously demanded scale.

5. The Gatekeepers Begin to Lose Their Moat

For decades, publishers, television networks, software companies, advertising agencies, and other institutions controlled scarce capabilities. They controlled the machinery required to produce and distribute professional-quality work. AI makes many of those capabilities increasingly reproducible by smaller players.

The institutions do not disappear. Their brands, networks, capital, relationships, and trust still matter. What weakens is their exclusive control over production capability. The moat begins to move from the ability to produce to the ability to attract attention, trust, distribution, and relationships.

6. From Mass Media to Micro-Media

Consider news and media. A conventional television operation requires reporters, editors, producers, studios, graphics, presenters, technical staff, and distribution infrastructure. AI and digital platforms can reproduce many of these functions at a fraction of the organizational scale.

At the same time, audiences are fragmenting. Instead of one dominant media organization serving everyone, thousands of smaller operations can serve specific interests, communities, languages, professions, or ideological and cultural niches. The result is not simply cheaper media. It is a lower economic threshold for becoming a media organization.

7. The Paradox: AI Powers Its Own Decentralization

There is an important paradox here. AI companies themselves require enormous amounts of capital, computing infrastructure, users, revenue, and feedback to develop increasingly capable systems. Their success depends on expanding adoption and usage.

But every improvement in AI capability gives users more power to operate independently of traditional organizations. The better the systems become, the more productive capability they distribute to individuals and small teams.

The technology therefore creates a feedback loop in which concentrated infrastructure can produce increasingly decentralized capability.

8. Concentration Above, Decentralization Below

This produces a structure that can look contradictory at first. Power may become more concentrated at the infrastructure layer, where a relatively small number of organizations control advanced models, chips, data centers, and computing capacity. Yet power can become more distributed at the application layer, where millions of individuals and small organizations gain access to capabilities that were previously expensive or institutionally scarce.

So AI can simultaneously concentrate infrastructure and decentralize productive capability.

9. From Wealth Transfer to Power Transfer

Most discussions of AI focus on wealth: who will become rich, who will lose income, which companies will dominate, and how profits will be distributed. Those questions matter, but there is a deeper question: who gains the ability to act? This connects directly to the question of who owns the value AI creates, explored in Universal Basic Equity: Who Owns the AI Future?

Historically, the pathway was roughly:

Capital → organization → capability → power

The emerging pathway may increasingly look like:

AI capability → individual leverage → power → potentially wealth

That is why the deeper disruption may not be a transfer of wealth at all. It may be a transfer of power—the ability to organize knowledge, create products, build media, launch services, and execute complex work without needing a large institution behind you.

Conclusion: When Power Leaks Out of Structures

The industrial revolution amplified physical labor. Computing amplified calculation. The internet amplified communication and distribution. AI may amplify the individual ability to organize knowledge and execute complex work. For the next stage of this argument, see Intelligence Is the Beginning, Not the Destination.

Large institutions will remain. But their size may become less tightly connected to productive capability than it was in the past. As capabilities become cheaper and more accessible, structures built around scarcity become less necessary.

The real disruption of AI may therefore be much larger than automation. It may be the gradual movement of productive power out of institutions and into individuals.

Power begins to leak out of structures and into people.