colorado AI law · SB 24-205 takes effect june 30, 2026 · 46 days · $20k/violation · run the 5-question check →
RIG. Get in touch
RIG
Capitulum I · Anno MMXXVI
✦

The patents & the doctrine.

A renaissance of the apparatus — three filings, two doctrines, one operator. The Deviation Engine. The 3D Lattice. PHRONEMA. Read the abstracts, the claims, and the math behind the median's collapse.

— I
M·R
the seal
— II
∴
deviation
— III
V
5 engines
— IV
∞
large numbers
— V
⊞
3D lattice
— VI
●
production
— filings · the receipts
RIG-DE-001
Deviation Engine · patent draft · 18 claims · in preparation
RIG-LAT-002
3D Governance Lattice · patent draft · 22 claims · in preparation
RIG-PHR-003
PHRONEMA tribal-knowledge vault · patent draft · 14 claims · in preparation

All three filed under non-exclusive perpetual license default. Inquiry reply within 24h.

— Plate I · the apparatus, photographed —

The lattice, as object.

The 588-cell governance lattice — rendered as a Da Vinci-style architectural model. Ivory cells are deterministic operators. Iron-black cells are probabilistic. Oxblood-glowing cores are deviation cells — the seam where every interesting thing happens. 7 × 12 × 7 = 588.

— fig. I · the lattice, turned588 cells · 15% deviation

The same lattice that holds three pending patents — the apparatus the studio runs on.

— Capitulum I.α —

The core math.

Robust-MAD-Z. SigmaScore. Escalation cost. The primitives every Deviate Engine inherits. Patent triggers at 5σ; deviation rises through Cognitive (±20σ), Nature (±20σ), and Physics (±30σ HARD).

— primitive · 01

Robust-MAD-Z

rawMAD = median(|scorei − median(scores)|)
Robust-MAD-Z = 0.6745 · (score − median(baseline)) / rawMAD

M-estimator on the deviation residual. The 0.6745 constant calibrates to a normal distribution. Robust = outlier-resistant. Z = unitless sigma score.

— primitive · 02

SigmaScore

SigmaScore = 0.3 · structural + 0.7 · behavior
Patent trigger ≥ 5σ

Behavior weighted heavier than structure — what the system does matters more than how it's drawn. Any output ≥ 5σ from baseline triggers patent-claim review.

— primitive · 03

Escalation cost

|n| ≤ 10 → 2|n|
10 < |n| ≤ 20 → 1024 · 3(n−10)
20 < |n| ≤ 30 → 1024 · 310 · 5(n−20)

Cost of deviation grows super-exponentially. Each sigma stage has a higher per-unit ceiling. Governance budget bounds where the engine can operate.

— gate logic

Three gates

  • Cognitive — block if any layer's Robust-MAD-Z < 3.0
  • Bell — block if |S| ≤ 2 (no entanglement signal)
  • Hawking — block if InternalVolume > 10·SurfaceArea (black-box)
  • Fine-tuning — block if ViableBand / ParameterSpace > 0.10
  • Vacuum — block if BaselineVariance < ε

The three-layer stack.

— Cognitive · engines 01–20 · ±20σ

Mechanism. Rupture. Collision.

Mechanism = 0.25·CausalLinkDensity + 0.20·FeedbackLoopClarity + 0.20·ConstraintVisibility + 0.15·FailureModeCoverage + 0.10·OperationalSpecificity + 0.10·Testability
Rupture = 0.25·AssumptionRemoval + 0.20·CategoryFrameBreak + 0.20·OrthodoxyViolation + 0.15·ReframeStrength + 0.10·FirstPrinciplesRebuild + 0.10·IrreversibilityOfInsight
Collision = 0.25·DomainDistance + 0.25·StructuralAlignment + 0.20·PracticalTransfer + 0.15·EmergentPropertyCreation + 0.15·MetaphorNonDecorativeness

Gate: any MADZ < 3.0 → BLOCK

— Nature · engines 21–30 · ±20σ

Pheromone. Flight. Spiral.

Ant pheromone: τ(t+1) = (1−ρ)τ + Σ τk
Lévy flight: P(s) ∝ s−1−α
Physarum flow: Q = (D/L)(pi − pj)
Firefly: β(r) = β0·e−γr²
Whale spiral: D′·ebl·cos(2πl)

Biology as architecture: every Nature-layer engine borrows a natural search heuristic — ant colony, Lévy, Physarum, firefly, whale optimization.

— Physics · engines 31–40 · ±30σ HARD

Fine-tuning. Bell. Hawking.

FineTuningRatio = ViableBand / ParameterSpace (BLOCK if > 0.10)
FineTuningEntropy = −Σ pi·log(pi)
BellParameterS = E(a,b) − E(a,b′) + E(a′,b) + E(a′,b′) (BLOCK if |S| ≤ 2)
Hawking gate: InternalVolume > 10·SurfaceArea → BLOCK
Vacuum gate: BaselineVariance < εvacuum → BLOCK

Hardest layer. Physics gates are the last filter before patent claim.

Representative Enhanced Brain (REB) · core formulas

Bayes update: P(H|E) = P(E|H)·P(H) / P(E)
FSRS retention: R(t,S) = (1 + t/(9S))−1
Ebbinghaus forgetting: R(t) = e−t/S
Expected value: EV = Σ Pi·Vi
Kelly criterion: f* = (bp − q) / b
Compound growth: A = P(1 + r)n
Power law: P(X ≥ x) ∼ x−α
KL divergence: DKL(P‖Q) = Σ P(x)·log(P(x)/Q(x))

Autonomous Digital Twin Brain (ADTB+) · formulas

Bayes: P(H|E) = P(E|H)·P(H) / P(E)
Shannon information: I(x) = −log P(x)
Mutual information: I(X;Y) = Σ P(x,y)·log(P(x,y)/(P(x)P(y)))
KL divergence: DKL(P‖Q) = Σ P(x)·log(P(x)/Q(x))
Cross entropy: H(p,q) = −Σ p(x)·log q(x)
Expected value: EV = Σ Pi·Vi
SIR diffusion: dI/dt = βSI − γI
Power law: P(X ≥ x) ∼ x−α
Wright's law: Y = a·Xb
Cumulative learning: output compounds with experience

Prediction Studio · scoring + calibration

Brier Score: BS = (1/N)·Σi(pi − oi)²
Log Score: LS = −(1/N)·Σi[oi·log(pi) + (1−oi)·log(1−pi)]
Confidence-weighted P: Pfinal = Σ(pi·ci) / Σci
Disagreement: σcouncil = √( (1/N) · Σi(pi − p̄)² )
Trimmed mean: Ptrim = mean( sort(probs)[k : N−k] )
Engine weight: wnew = (1−η)·wold + η·reward
Market-blended: forecast = swarm·agent + market·prior
EVI: E[utility|collect] − E[utility|no] − cost

— targets: Brier < 0.15 · Log Loss < 0.40 · ECE < 0.05/bucket · feedback latency < 1hr

Public surface only. Full claim language, parameter values, and apparatus details are under filing (RIG-DE-001 · Q4 2025). Patent trigger: SigmaScore ≥ 5σ on Robust-MAD-Z.

— Capitulum I —

The Deviation Engine.

RIG-DE-001 · patent draft · in preparation — the apparatus that samples the seam between deterministic and probabilistic flow.

FIG. I — apparatusRIG-DE-001
          ╔═════════════════════════╗
          ║  probabilistic ceiling  ║   ← variable outcomes
          ╚════════════╤════════════╝
                       │   sampled
          ┌───────────  ▼  ───────────┐
          │    deviation engine       │   ← ±30σ ladder
          │    (apparatus, patent)    │
          └───────────  ▲  ───────────┘
                       │   measured
          ╔════════════╧════════════╗
          ║  deterministic floor    ║   ← fixed inputs
          ╚═════════════════════════╝
Fig. I — the seam between floor and ceiling, sampled by the apparatus on the ±30σ ladder.
RIG-DE-001 · patent draft · Q4 2025

The Deviation Engine.

patent draft in preparation inventor · Rodgers

An apparatus and method for controlled deviation across the deterministic↔probabilistic spectrum. The engine samples a deterministic floor of fixed-input outputs and a probabilistic ceiling of sampled outputs; it then selects a deviation magnitude on a bounded ±Nσ ladder under a governance budget, surfacing signal the median refuses to produce.

selected claims
  1. A method comprising sampling a deterministic floor and a probabilistic ceiling, and selecting an output by deviation against a ±Nσ ladder.
  2. The method of claim 1 wherein N is bounded by a deviation budget enforced by a governance layer.
  3. The method of claim 1 further comprising a feedback loop wherein deviation outputs are scored against a reference distribution and the budget is updated.
  4. The method of claim 1 wherein the deviation magnitude is selected by an adversarial simulator coordinating multiple engines across the ladder.
core formula — deviation output
Di = Fi + σi·ξi,   ξi ~ 𝒩(0, 1),   σi ∈ [−30, +30]
governance budget — bounds deviation across cells
Σi |σi| ≤ Brow · Ncells
— Capitulum II —

The 3D Lattice.

RIG-LAT-002 · patent draft · in preparation — 3,087 cells across 7 rows · 21 columns · 21 depth. Every engine inherits the topology.

FIG. II — topologyRIG-LAT-002
              7 rows · 21 cols · 21 depth
              ═════════════════════════
              ┌────┬────┬────┬────┬────┐
              │ R₁ │ R₂ │ R₃ │ R₄ │ R₅ │   ← capability spine
              ├────┼────┼────┼────┼────┤
              │ ▒▒ │ ░░ │ ▓▓ │ ░░ │ ▒▒ │
              ├────┼────┼────┼────┼────┤
              │ ▓▓ │ ▒▒ │ ░░ │ ▓▓ │ ░░ │
              └────┴────┴────┴────┴────┘
                ▒▒ deterministic  ░░ probabilistic
                       ▓▓ deviation
              3,087-cell lattice — public on GitHub
Fig. II — each cell holds three operators (det, prob, dev) and a row/col/depth governance index.
RIG-LAT-002 · patent draft · Q4 2025

The 3D governed lattice.

patent draftin preparation3,087 cells

A three-dimensional lattice topology for governed agentic systems. Each cell holds a deterministic operator, a probabilistic operator, and a deviation operator that arbitrates between them under a row/column/depth governance budget. Cells communicate via deviation bridges scored on the ±30σ ladder.

selected claims
  1. A three-dimensional lattice of computational cells, where each cell encodes a deterministic↔probabilistic spectrum position.
  2. The lattice of claim 1 wherein governance is enforced by row, column, and depth indices independently.
  3. The lattice of claim 1 wherein cells communicate via deviation bridges scored on the ±30σ ladder.
  4. The lattice of claim 1 wherein the row dimension corresponds to a capability spine across operational discipline, innovation architecture, sales, AI systems, IP creation, and polymath range.
cell output at lattice position (r, c, d)
Lr,c,d = wdet·Fr,c,d + wprob·Pr,c,d + wdev·Dr,c,d
where w_det + w_prob + w_dev = 1 (convex combination), and r ∈ [1,7], c ∈ [1,21], d ∈ [1,21].
— Capitulum III —

PHRONEMA · the vault.

RIG-PHR-003 · patent draft · in preparation — the library-of-Alexandria-for-one, with heuristic embedding bridges that extend operator trajectory.

FIG. III — flowRIG-PHR-003
       ┌───────────────────────────────────────┐
       │   web · device · interaction streams  │
       └────────────────────┬──────────────────┘
                            │
                  ┌─────────▼─────────┐
                  │      HEB++        │   indexing
                  └─────────┬─────────┘
                            │
                  ┌─────────▼─────────┐
                  │     PHRONEMA      │   trajectory-extending
                  └─────────┬─────────┘
                            │
                       ◀───┤   ◯  operator
                            └───▶
Fig. III — capture, index, retrieve. The retrieval is biased toward extending operator trajectory.
RIG-PHR-003 · patent draft · Q1 2026

PHRONEMA · HEB++.

patent draftin preparationtribal vault

A personalized continuous-capture knowledge vault with heuristic embedding bridges. The system captures web, device, and interaction streams; indexes them via heuristic embedding bridges (HEB++); and retrieves memories under a trajectory-extension bias that prefers extending the operator's future reach over reproducing prior actions.

selected claims
  1. A continuous-capture system producing operator-specific embeddings indexed by a trajectory model.
  2. The system of claim 1 wherein retrieval is biased toward extending operator trajectory rather than reproducing prior actions.
  3. The system of claim 1 further comprising heuristic embedding bridges across modalities (text, image, audio, telemetry).
  4. The system of claim 1 wherein the trajectory model is updated by operator-confirmed retrievals as positive samples.
retrieval score for memory m, given query q, trajectory T
R(m|q, T) = ⟨em, eq⟩ · 𝟙(m extends T) · τ−Δt
— Capitulum IV —

The law & the creative booster.

Why the math of LLMs is collapsing creativity — and what the apparatus does about it.

The Law of Large Numbers, restated for LLMs.

The Law of Large Numbers says: as you draw more samples from a distribution, the sample mean converges on the true mean. That's the math.

The training-set version: as you collect more text, the model's outputs converge on the median of human writing. Useful for ground truth. Devastating for creativity.

What it looks like: a tree with one trunk and no branches. Useful + correct + average. Year over year, every model converges on the same voice, the same answer, the same conclusion.

The Creative Booster reinstates novelty by measuring how far an output sits from the training-data median, then injecting governed deviation. But only at the right moment, and only in the form of questions, not statements.

Questions extend trajectory. Statements collapse it. The Booster scores both — and rewards the question form when novelty matters.

novelty score N
N = |y − μ|σtrain · α2 · 𝟙(y ∈ question-space)
creativity / novelty training samples → median collapse creative booster — question injections ? ?
— median collapse · LLN at work ↑ question injection · α · 𝟙(question)
— Capitulum V —

Forty engines. One ladder.

A sample of the 40 deviation engines running on the ±30σ adversarial benchmark grid. Production engines ship; staging engines compete; bench engines wait.

I
creative-booster
novelty injection · question-form indicator · α-weighted
production
II
med-triage
clinical encounter routing · 200M+ events
production
III
field-commander
rural network operator agent · offline-first
production
IV
patent-prior-art
prior-art scan on USPTO + Google Scholar
production
V
sigma-bench
live ±Nσ benchmark plotting
production
VI
song-spine
Larry Studio · arrangement skeleton
production
VII
lyric-deviator
lyric novelty score · median-aware
production
VIII
seam-finder
locates the deterministic↔probabilistic seam
production
IX
governance-budget
enforces ±Nσ deviation budget across cells
production
X
noise-floor
detects when σ collapses below signal threshold
staging
XI
soul-drift
operator soul-stack drift detector
staging
XII
showup-detector
flags when the operator hasn't shown up
bench

— 28 more engines in production · 8 staging · 4 on the bench · see the full engine list — σ + score + budget — inside the RIG build.

— Capitulum VI —

The research & development arc.

How the doctrine got built — from cornfield to cluster to patent filings.

2009

Army CI

Six years of counterintelligence operations. The discipline layer. Where "if you haven't built it yourself" became axiomatic.

2015

Oracle · M&A + Innovation Architecture

$35M opex reductions. Founded + scaled an innovation subsidiary with 30 headcount and $27M budgets.

2020

Emopti · ED Triage at Scale

ED wait times: 3hr → 15min across 5+ hospital systems. $10M Series A secured during COVID. 200M+ encounters routed by the first iteration of what became the Womb-Named engine.

2024

The Home Rack · Online

5× GPU + 3× Mac Studio assembled in the Denver basement. First self-funded inference cluster. The flag day for ship-on-the-rack discipline.

2025

RIG · Studio Founded

Rodgers Intelligence Group. Business Intelligence Studio with 24 systems. Deviation Engine doctrine formalized. First two patent drafts in preparation.

2026

PHRONEMA drafted · Vol. IX launched

Third patent draft in preparation. Public site, manifesto, and the 100-day window become the offer.