sovereign-system-spec

Architectural patterns and terminologies for sovereign AI systems. Eliminating the Prose Tax and reclaiming intellectual provenance through local-first engineering constraints.

View the Project on GitHub kenwalger/sovereign-system-spec


layout: default title: Prose Tax term_name: Prose Tax term_description: The financial and computational premium paid to process conversational boilerplate, formatting, and non-essential semantic structure within AI systems. ————————————————————————————————————————————————————————-

Prose Tax

Definition

Prose Tax is the financial and computational premium paid to process conversational boilerplate, formatting, and non-essential semantic structure within AI systems.

It represents the hidden operational cost incurred when machines are required to parse language optimized for human readability rather than computational efficiency.

Within Sovereign Systems, Prose Tax is treated as a measurable architectural cost rather than an unavoidable property of language.

Origin

The term Prose Tax was first formalized as part of the Sovereign Systems Specification by Ken W. Alger in 2026.

Why It Matters

Large language models consume tokens.

Every token consumes:

While natural language is highly effective for human communication, many AI systems spend significant resources processing conversational filler, formatting conventions, pleasantries, redundant explanations, and other low-information content.

As systems scale, this overhead becomes economically significant.

Prose Tax is particularly visible in:

Example

Consider two equivalent payloads:

Human-Oriented

Hello! I hope you're doing well today.

Could you please review the attached support ticket and determine whether the issue has been resolved? If possible, provide a brief summary of your findings.

Thank you very much.

Machine-Oriented

{
  "task": "review_ticket",
  "ticket_id": "12345",
  "action": "determine_resolution_status",
  "return": "summary"
}

Both communicate the same intent.

The first requires significantly more token processing while providing little additional operational value to the machine.

The difference represents a form of Prose Tax.

Relationship to Context Tax

Prose Tax and Context Tax are closely related but distinct.

Prose Tax measures the cost of processing unnecessary language.

Context Tax measures the cost of processing excessive or weakly relevant information.

Reducing Prose Tax often reduces Context Tax, but the two concepts are not interchangeable.

The Sovereign Approach

Sovereign Systems seek to minimize Prose Tax by:

The objective is not to eliminate natural language.

The objective is to reserve natural language for situations where it creates meaningful value.

References