PATENT PENDING — US #64/036,090

Technical Overview

RNDA is a data architecture protocol, not a product. This page covers the mathematical foundations, empirical benchmarks, and architectural claims. For engineers and researchers evaluating RNDA for licensing or integration.

Core Architectural Principle

"Uncompressed data is not a state that exists in the RNDA system — not during storage, transit, computation, or output. Input data is encoded to pattern signatures, the raw data is permanently discarded, and contextually appropriate outputs are reconstructed on demand from signature overlap. Reconstruction is generative, not retrieval."

This inverts the fundamental assumption of every existing data system. Traditional systems treat the uncompressed form as the canonical source of truth. RNDA has no canonical uncompressed form — the pattern signature store IS the source of truth, and outputs are always reconstructed derivatives.

Mathematical Foundation

Sparse Distributed Representations (SDR)

Input data of arbitrary size is encoded to a binary vector of dimension N=2048 where approximately w=41 bits are active (2% density). The encoding function E: Input → {0,1}^2048 satisfies three properties:

Deterministic

Same input always produces the same SDR signature

Similarity-preserving

Semantically similar inputs produce overlapping bit patterns

One-way

Given a signature, recovering the input requires brute-force over C(2048,41) ≈ 10^72 possibilities

Vector Symbolic Architecture (VSA)

RNDA uses VSA binding operations to encode structured relational information without retaining the underlying data. A binding operation ⊗ combines multiple entity encodings into a single signature:

sig = encode(agent) ⊗ encode(action) ⊗ encode(target) ⊗ encode(time)
# Stores the relationship — original entities discarded

query_result ≈ sig ⊗ encode(action) ⊗ encode(target) ⊗ encode(time)
# Retrieves approximate agent encoding — NOT original text

Compression Scaling (Empirical)

The signature is always 256 bytes regardless of input size. Compression ratio = input_size / 256. Linear scaling confirmed on real Wikipedia data:

Document Size	Raw Size	Signature	Compression	Reduction
~1 KB	986 bytes	256 bytes	4x	74%
~5 KB	4,545 bytes	256 bytes	18x	94%
~20 KB	16,482 bytes	256 bytes	64x	98%
~50 KB	40,966 bytes	256 bytes	160x	99.4%
1 MB	1,000,000 bytes	256 bytes	3,906x	99.97%

Scale Benchmarks

Tested on Simple English Wikipedia — 10,000 real articles. All raw data permanently discarded after encoding.

Articles Encoded

10,000

Real Wikipedia articles

Raw Data Stored

NONE

Permanently discarded

Semantic Similarity

0.547

Improves with corpus size

Query Latency

68ms

Across 10K signatures

Query Results (10,000 articles)

World War II military historyOperation Overlord0.538

Albert Einstein physics relativityRelativity0.579

Climate change global warmingGreenhouse effect0.594

Olympic games sports athletesSummer Olympic Games0.609

Shakespeare English literatureLove's Labour's Lost0.639

Context-Dependent Reconstruction

The same signature store produces different valid outputs for different query contexts. Neither output is the original data — both are generated reconstructions appropriate to the context.

Query: "How do governments respond to environmental challenges?"

CONTEXT: ACADEMIC

"Governmental response operates through a multifaceted framework encompassing resource management and stewardship mechanisms, with policy frameworks accounting for biological and ecosystem dynamics."

CONTEXT: EXECUTIVE

"Two primary mechanisms: resource management leveraging natural resources for sustainable solutions, and product stewardship frameworks. Ecological yield ties environmental protection to economic sustainability."

↑ Same signature store. Different outputs. Original data: PERMANENTLY DISCARDED.

Patent Claims Summary

US Provisional Patent Application #64/036,090 covers 26 claims across method, system, and apparatus categories. Key independent claims:

Claim 1

Core Method

A method wherein input data is encoded to SDR signatures, raw data is permanently discarded, and contextually appropriate outputs are generated from signature overlap without retrieving the input data.

Claim 2

System

A system comprising an encoding module that discards input data after encoding, a signature store containing no raw data, and a reconstruction engine generating different outputs per query context.

Claim 5

Context-Dependent Reconstruction

A method wherein the same signature store produces different valid outputs for different context parameters, and no output constitutes retrieval of original input data.

Claim 22

AI Infrastructure Cost Reduction

A system reducing AI infrastructure costs by at least 10x by storing only SDR signatures of training data, model weights, and user data without retaining uncompressed forms.

Claim 23

Behavioral Identity Preservation

A method enabling interaction with a human behavioral identity model after the subject is no longer able to participate, without any stored record of original behavioral events.

API Reference

The RNDA demo API is available at api.rnda.io. Rate limited to 5 queries per IP per 24 hours. Contact us for expanded access.

POST/api/encode

Encode text to SDR signature. Raw text permanently discarded.

{ "text": "string", "label": "string (optional)" }

POST/api/query

Query signatures, generate 3 contextual reconstructions.

{ "query": "string", "contexts": ["academic", "general", "executive"] }

GET/api/health

Health check + signature count.

GET/api/rate-limit

Check remaining queries for your IP.

Licensing & Integration

RNDA is available for licensing. Whether you're building AI infrastructure, training systems, or privacy-first applications — contact us to discuss integration.

Contact for Licensing →

US Patent Application #64/036,090 | Priority Date April 11, 2026 | ZiggyTech Ventures LLC