For years, companies have tried to solve the same organizational problem: how do you reduce friction, remove bureaucracy, and get valuable work into the hands of customers faster?
Amazon had the two-pizza team: no team should be so large that two pizzas could not feed it. Then came the idea of single-threaded teams: one team, one mission, one leader, one clear problem to solve. Netflix and others pushed versions of pods, squads, and cross-functional cells.
The names differ, but the intent is the same: smaller teams have fewer handoffs, fewer dependencies, and less room for responsibility to disappear into the group. Companies keep trying to create smaller units of execution because big organizations naturally create drag. Anyone who has worked inside a large company knows how this happens. Product sits in one place. Engineering sits somewhere else. Data has its own queue. Legal needs to review. Marketing needs a brief. Finance needs a business case. Leadership needs alignment. By the time an idea becomes something a customer can actually touch, half the energy has leaked out of the system.
AI changes the shape of this problem.
AI makes existing teams more productive. That is true, but it is probably the least interesting part. The bigger shift is that AI allows one capable person to do work that previously required a small team. One person can research a market, analyze customer feedback, draft a product concept, build a prototype, test it, review results, and iterate. Not perfectly. Not always. Not in every domain. But well enough and often enough, that the unit of value creation inside companies is going to shrink again: From department, to team, to squad, to pod, and now, in some cases, to one person.
This is the rise of the One-Person Value Factory.
By that, I do not mean a freelancer inside the company. I also do not mean a heroic individual bypassing the organization, or someone doing ten jobs badly. A One-Person Value Factory is a person with enough context, judgment, tools, and authority to take a problem from insight to shipped value with minimal dependency on the machinery around them (the key phrase being shipped value).
AI is going to create a lot of activity that looks like progress: More documents, more analysis, more dashboards, more content. But none of that matters unless it reaches a customer, improves the business, removes friction, saves money, creates revenue, or changes behavior. The One-Person Value Factory is not measured by how much it produces. It is measured by how quickly it turns ambiguity into useful output.
That has real implications for organizational design. The old model assumes that execution requires coordination across specialties. The One-Person Value Factory model will increasingly assume that execution starts with autonomous value creators, supported by specialists only when needed. Instead of assembling a team around every idea, companies will ask a different question: can one strong operator take this far enough before we involve the machine?
This is where management will need to change. Managers will not just allocate people to projects. They will need to decide where autonomy is safe, where synchronization is necessary, and where control is slowing the company down. And this is where the hard part begins.
AI reduces production friction. It does not remove organizational friction. In fact, it may make synchronization harder. If ten people can each move five times faster, the organization does not automatically become fifty times faster. It may simply become ten fast-moving parts creating confusion in parallel. That confusion can show up as duplicate work, inconsistent customer messages, conflicting experiments, brand drift, security gaps, compliance risk, and customizations that do not add up to company progress. The bottleneck will move: It used to be production, increasingly, it will be synchronization.
I believe this to be the next organizational challenge. How do you let One-Person Value Factories move fast without turning the company into a bag of disconnected experiments? You certainly don’t slow them down. The answer is to be much clearer on the few things that must be synchronized. Strategy must be synchronized, so people know which problems matter. Standards must be synchronized, so people understand the guardrails around architecture, quality, risk, data, compliance, and brand. Customer experience must be synchronized, so the company does not feel like a collection of unrelated products.
But the work itself should not be over-synchronized. In my opinion, this is where many companies will get it wrong. They will see the risk of AI-enabled autonomy and respond with committees, approval flows, and more governance. They will take a technology that can compress weeks into days and wrap it in a process that turns days back into weeks, if not months. They will take the Lamborghini and hitch it to a horse-drawn carriage.
We are living through an era where time has become even less neutral. Decisions require movement before everyone is comfortable. They require fast diagnosis, fast action, fast feedback, and fast correction. You do not get to wait until the structure is perfect. You need traction while the structure is still messy. That makes the One-Person Value Factory especially powerful.
The question becomes: who in the organization can be trusted to own a problem end to end?
Those people will become disproportionately valuable. Not because they “know how to use AI.” That will become table stakes. They will be valuable because they combine business judgment, customer understanding, execution discipline, and enough technical fluency to move without constant translation.
The organizational question is no longer only “how do we build better teams?” It is also “how do we design a company where one person can create meaningful value without being trapped by the machinery around them?”
That is not a small shift.
The company of the future may still have teams, pods, functions, and leaders. But the atomic unit of progress will get smaller. And in many cases, it will be one person with context, judgment, AI leverage, and permission to move.
That is the One-Person Value Factory: not a person doing everything, but a person able to move value through the system before the system slows it down.
Turning Stuff Around 