I sketched the first version of this on my phone at 33,000 feet, using ZAMARA, the multi-model agent system I run for myself, as a thinking partner. The question I kept circling was simple: why do two companies buy the same models, hire similar people, and end up worlds apart on what AI actually does for them?

The answer isn’t the technology. It’s maturity. Organizations adopting AI move through four distinct stages, and the gap between them is organizational, not technical. PwC’s research put a number on the spread: roughly 74% of AI’s value is concentrated in the 20% of companies that lead, a performance gap of about 7.2× between the leaders and everyone else. That gap is a maturity gap.

In brief
What’s changing
AI capability is now a commodity; the differentiator is how far up the maturity curve your organization has actually climbed.
Why it matters
Most organizations are stuck at Stage 2: faster individuals, unchanged results. The value lives at Stages 3 and 4.
What to reconsider
Stop measuring tool adoption. Start measuring whether work itself has been redesigned around human judgment and agentic execution.
  1. Stage 01

    Awareness

    AI is understood, not yet used. Curiosity outruns capability.

  2. Stage 02

    Productivity

    Individuals get faster, but value stays trapped at the desk. The Stage 2 Ceiling.

  3. Stage 03

    Context to Intelligence

    The organization feeds AI its own context. Work is redesigned, not just tasks.

  4. Stage 04

    Humangentic

    Human judgment and agentic execution compound. The organization runs on both.

Stage 1: Awareness

Everyone has heard of AI; few have used it for anything that matters. Leadership talks about it in all-hands meetings, a pilot or two runs in a corner, and the dominant emotion is anxious curiosity. Awareness is necessary and cheap, and it produces almost no value on its own. The risk here is mistaking conversation for capability.

Stage 2: Productivity

Individuals get faster. Someone drafts emails in half the time; an analyst summarizes a report in seconds. It feels like progress, and it is, at the desk. But the value stays trapped there. The organization is a collection of slightly-faster people doing the same work in the same shape, and the P&L doesn’t move. This is the Stage 2 Ceiling, and it is where the large majority of organizations quietly stall. Faster horses, not cars.

Stage 3: Context to Intelligence

The climb out of Stage 2 begins when the organization stops feeding AI generic prompts and starts feeding it its own context: its data, its decisions, its institutional knowledge. This is where work gets redesigned rather than merely accelerated. It is unglamorous and it is where the compounding starts: retrieval over your own corpus, judgment encoded into where AI is trusted and where it isn’t, processes rebuilt around what the machine is actually good at.

Stage 4: Humangentic

A humangentic organization is one where human judgment and agentic execution compound instead of compete. Humans set direction, exercise judgment, and own accountability; agents carry the execution, at machine speed and scale.

This is the destination. Not “AI replaces people” and not “people ignore AI,” but a genuine division of labor: the human owns the why and the should-we; the agent owns the do. The judgment layer is the human contribution, and it is the thing that becomes scarce, and therefore valuable, as execution gets cheap.

What to take into the room

  • Ask which stage you are actually at, measured by redesigned work, not tools deployed.
  • If you are at Stage 2, name it. The ceiling is the most expensive place to be comfortable.
  • Fund Stage 3, context to intelligence. It is where the value gap opens.
  • Design the judgment layer deliberately: decide, per decision, whether it is fully delegated to an agent, kept human-in-the-loop, or never automated.

This is a living framework. It underpins everything else published here: the Stage 2 Ceiling, graded governance, and the anatomy of an agent are all views into this same climb.