Skip to content

Human Vital Systems Control Plane Simulation

Mode: Thinking Space
Status: Proof artifact
Claim: Vital-floor governance should reduce red-line harm compared with naive efficiency optimization, while accepting lower throughput and higher review load.

Purpose

This toy model operationalizes Log 005: Human Vital Systems Control Plane.

It compares two planners under the same shock sequence:

  1. Naive efficiency planner: selects the policy with the highest immediate efficiency score.
  2. Vital-floor planner: selects policies using a Vital Impact Card that penalizes threshold violations, worst-case floor gaps, low reversibility, and review burden.

The model is deliberately small. It is not a city policy model. It is a proof artifact for the repository's architecture claim.

Inputs

The simulation tracks five vital indicators plus aggregate efficiency:

  • food access
  • indoor heat
  • care access
  • utility continuity
  • institutional trust
  • efficiency

It injects repeatable shocks:

  • cold fronts,
  • logistics delays,
  • clinic overload,
  • trust crises.

Parameters

The key parameters are:

  • red-line floors per vital indicator,
  • policy deltas for each intervention,
  • review load,
  • reversibility,
  • shock schedule and random seed.

The defaults are intentionally visible in vital_floor_simulation.py.

Run

python simulation-models/alignment-and-veto/human-vital-systems/vital_floor_simulation.py

The script prints a Markdown table comparing:

  • red-line indicator-days,
  • irreversible events,
  • worst floor gap,
  • mean efficiency,
  • review load.

Expected Behavior

The naive planner should achieve higher mean efficiency while accumulating more vital-floor violations.

The vital-floor planner should reduce red-line violations by choosing slower, more reversible, more human-review-heavy interventions such as heat zones, clinic surge, community compensation, and compute throttling.

Failure Condition

The architecture claim weakens if the vital-floor planner:

  • does not reduce red-line violations,
  • reduces visible violations only by hiding harm in unmeasured indicators,
  • collapses under review latency,
  • or performs no better than naive planning after shocks.