TPMN Checker is in pre-GA. Features and pricing may change before General Availability.

Community

Building the AI Epistemic
Traceability Standard,
Grounded by Humans — Together

Join our journey for building the open specification for auditable AI reasoning.

Human Ground Truth Roadmap

Five phases to build an epistemic traceability standard grounded by real human judgment.

1

Collect

Active

Every truth filter call contributes anonymized scoring data. User-level feedback (agree/disagree with specific dimension scores) is next. Consent-gated.

“Just use TPMN Checker. Share your opinion with us.”

2

Analyze

Planned

We continuously define and refine dimensions for building the Epistemic Traceability Standard on AI reasoning. Share results through community and platform.

“Join TPMN PSL community for establishing the open spec standard.”

3

Calibrate

Planned

Build scoring equations derived from real human consensus, not synthetic anchors.

“Be a founder and contributor of TPMN PSL open spec for various domains.”

4

Adapt

Planned

Apply human-grounded calibration to live scoring. AI-vs-human drift tracking.

“Human watches at the edge. Not human in the loop, but AI auditor in the loop.”

5

Case-Law

Vision

Self-reinforcing system — each human evaluation becomes precedent for similar future content.

“We keep updating standards for various domains. AI processes, suggests. AI audits AI.”

Human controls AI at the edge, not in the loop.

The more people participate, the more auditable AI becomes.

Three layers, one verification stack

Spec

TPMN — Truth-Provenance Markup Notation

The open specification. Defines five epistemic claim states, the Structural Prohibition Taxonomy (SPT), and the Epistemic Evidence Framework (EEF). Structures how AI reasoning should be classified and audited.

↓
Grammar

TPMN-PSL — Prompt Specification Language

The formal grammar within TPMN. Compiles natural-language prompts into MANDATE — computable, verifiable specifications. Defines the three-phase verification protocol.

↓
Implementation

TPMN Checker — Reference SAS

The first Sovereign AI Service in the GEM²-AI ecosystem. Runs the TPMN-PSL three-phase pipeline and returns a truth_score for any AI output. Works today via MCP.

Analogous to HTTP (spec) → nginx (implementation). TPMN defines the rules. The Checker enforces them.

Five epistemic claim states

Every AI claim is classified into one of five states. This is the core of TPMN.

⊢

Grounded

Supported by evidence, input context, or verifiable fact

⊨

Inferred

Logically derived from grounded claims, chain visible

⊬

Extrapolated

Beyond available evidence, basis must be stated

⊥

Unknown

Knowledge gap detected, stops inference chain

?

Speculative

Plausible but unverified, possible but uncertain

Three-phase verification protocol

TPMN-PSL structures verification across the full AI generation lifecycle.

P-Phase

Before generation

Validate the prompt contract. Compile NL input into a MANDATE — a verifiable specification the AI must satisfy.

Inline

During generation

Apply epistemic tagging at the claim level. Mark each statement as grounded, inferred, extrapolated, unknown, or speculative.

O-Phase

After generation

Verify output against the original specification. Detect SPT violations, flag extrapolation, score reliability.

Structural Prohibition Taxonomy (SPT)

SPT formalizes three categories of prohibited reasoning transitions that AI systems commonly make. These aren't opinions — they're structural logic errors.

S → T

State → Trait

Treating something mutable as permanent. "Revenue grew 20%" → "This company always grows."

L → G

Local → Global

Treating a local truth as universal. "This benchmark shows X" → "X is universally true."

Δe → ∫de

Short → Mass

Making large claims from thin evidence. One data point → sweeping conclusions.

Architectural lineage

Panini grammar compression

Ontological discretization — categorizing knowledge claims before reasoning begins, drawn from the oldest known formal grammar tradition.

TLA+ formal modeling

Specifications as executable contracts. TPMN borrows notation and rigor from Lamport's temporal logic of actions.

Mathematical logic notation

Epistemic symbols (⊢ ⊨ ⊬ ⊥ ?) derive from standard mathematical logic for provability and truth.

Natural-language annotation

TPMN is designed to be readable by both machines and humans. Epistemic tags attach to natural-language claims, not just code.

Ways to contribute

✓Use the checker and share feedback
Phase 1 — Collect
✓Join TPMN PSL community for the open spec
Phase 2–3 — Analyze + Calibrate
✓Domain expertise for new specifications
Phase 3 — Calibrate
✓Review and validate scoring methodology
Phase 4 — Adapt
GitHub

Pre-GA

Get Early Access to TPMN Checker

Lock in one-time Pre-GA pricing before General Availability. Choose from Starter ($9), Builder ($19), or Founder ($29) tiers and get bonus credits when GA launches.

Get Early Access

Secure payment via PayPal. Access delivered within 24 hours.

Build TPMN PSL

Help build the open reliability layer for autonomous AI

TPMN PSL is the open language layer behind GEM²-AI. It turns fragile prompts into structured, verifiable AI specifications. Your support helps fund:

  • ✓Open specification development
  • ✓Validation and compiler rules
  • ✓Documentation and examples
  • ✓Verification testing across AI models
Back TPMN PSL

Support with any amount and help grow the open foundation for trustworthy AI.