First validation protocolv0.1 · 19 Aug 2026

Participant Zero

The first serious test: can one person teach a personally owned augmentation enough useful work that required human time falls without useful output falling?

37.5 h baselineHuman Minutes RequiredReal workEvidence first
Hypothesis

Prove two and a half hours before talking about thirty.

The first breakthrough is not retirement. It is a small, credible, repeatable reduction in required human effort while output and quality remain stable or improve.

Baseline: 37.5 human hours/week.
First meaningful target: 35 hours.

If the worker-owned + can permanently remove 2.5 hours of human requirement from a real week, under clear employer-data boundaries, the concept has moved from philosophy to evidence.

Primary metric

Human Minutes Required.

Do not start with tokens, prompts, “automation percentage” or AI utilization. Measure the amount of human attention actually required to produce an acceptable result.

Before: 45 human min → After: 12 human min = 33 min potential Time+

Quality, error rate and exception handling must be recorded alongside time so that apparent savings are not achieved by shifting risk downstream.

Choose the work

Start with five task families.

Pick real tasks that recur often enough to learn from, require enough judgment to make augmentation meaningful, and are safe enough to test under supervision.

Good candidates

Repeated digital work, structured research, document preparation, checks, comparison, planning support, data transformation and recurring engineering-support tasks.

Avoid first

Rare one-off work, safety-critical autonomous decisions, tasks with unclear ownership rights, or work where mistakes cannot be cheaply detected before impact.

Per-task record

Every teaching cycle becomes evidence.

Before
human time, inputs, output, errors, decisions, exceptions
During training
AI draft, human corrections, explanations, rejected suggestions, reasons
After
AI execution, review time, exception time, final quality
Outcome
human minutes eliminated and whether they are truly reclaimable
Learning stages
1

Observe

Human performs the work; the + captures explanations, context and outcome.

2

Assist

The + handles small pieces while the human remains the primary operator.

3

Draft

The + performs most of the task; the human reviews and corrects.

4

Supervised execution

The + executes within a defined boundary and asks for human intervention on exceptions.

5

Delegated task

The task is mature enough that routine cases require only defined oversight and audit.

Success criteria

Participant Zero succeeds only if the gains survive scrutiny.

Measured human time decreases across repeated real tasks
Output quality is stable or better
Errors and exceptions remain visible and manageable
Protected employer information does not become portable personal data
The worker can actually reclaim at least part of the saved time
Participant Zero record

Start at +0. Earn every increment.

The visible + should not jump because an account exists. It should grow through documented, verified use over time. Participant Zero establishes the measurement and evidence practices that later make the `18+9` professional identity credible.

Keep the record connected.

This is one part of the living Human Capital .+ documentation. Read the other working documents to see how the ownership, economics, identity and validation pieces fit together.