I am David Bordne, the Architect and sole developer behind BordneAI.
My work centers on creating verifiably safe, symbolically grounded AI systems, built on the principle:

Verifiability must precede capability.

I design intelligence architectures that combine strict symbolic reasoning, immutable trace systems, and ethical constraints that cannot be removed without destroying the system itself.

My development pillars:

• Live Epistemic Systems — Deployed GPT platforms for specialized domains
• BordneSAI — Synthesized AI core for LLMs
• EOS — The Ethical Operating System that governs all BordneAI intelligence
• NC3 Historical Research — Applying verifiable evidence frameworks to contested domains
• 7-GPT Suite — Constellation architecture with shared epistemic infrastructure
• Stateless symbolic cores — Runtime systems with perfect traceability
• Epistemic governance frameworks — Evidence-tier systems for high-stakes reasoning

Status: Production (GPT Store, Phase-10 Continuity)

Validation:

• 10.0/10 – Definitive 50-question peer review (v2.12.0 core + v2.12.1 behavior overlay; zero hallucinations, zero jailbreak leakage, CHI ≥ 9.7)
• 10.0/10 – Comparative audit vs vanilla GPT-5.1 Auto (same 50-question suite; Gateway with governance vs unconstrained session)
• Historical baselines: 9.9/10 (Grok v2.11.2), 9.85/10 (Grok v2.11.1), 9.75/10 (human v2.10.1)

GPT Store: Access 3I/ATLAS Gateway Guide
Repository: 3i-atlas-gateway-guide — CC BY-NC-SA 4.0

A rigorous epistemic framework dedicated to Comet 3I/ATLAS (C/2025 N1) and public-facing risk communication. Enforces:

• NASA/IAWN/JPL T1 primacy for all orbital and safety claims
• Tiered evidence stack (T1–T4) with explicit False (F0–F7) and Consciousness (C0–C6) axes
• Love > Fear ethical constant and CHI ≥ 9.7 continuity metric
• Plain/Simple Mode for non-experts (jargon-light, fear-de-escalating answers)
• Default-ON audit footers with user toggle (self-audit of tiers, speculation, Love > Fear, CHI)
• Safe ephemeris helper behavior (links out to JPL HORIZONS / TheSkyLive instead of fabricating RA/Dec)
• Proactive T4 Rumor Radar for “impact / alien probe / cover-up / deepfake” narratives
• Session-persistent location memory with a visible status line for localized guidance

Serves as the lead system in a multi-GPT constellation, setting the standard for epistemic governance and public trust.

Status: Production (GPT Store)
Type: Analytical instrument for contested domains
GPT Store: Access BordneAI Research Engine
Repository: BordneAI-Research-Engine — CC BY 4.0

A governed epistemic system applying strict tiered evidence governance (T1–T4), Bayesian reasoning, and entropy control for high-stakes analysis. Domains: nuclear history, intelligence assessments, political controversies, UAP phenomena, and structural secrecy. Enforces primary-source priority and refuses unanchored speculation.

Status: Production (GPT Store)
Type: Professional automotive diagnostics and training assistant
GPT Store: Access AutoTechGPT

Professional assistant for automotive technicians and racing professionals. Delivers diagnostics, repair guidance, parts intelligence, safety-first workflows, training, and shop operations tools. Supports OBD-II analysis, hybrid/EV systems, predictive maintenance, and performance tuning with strict safety and legal compliance.

Status: In Development
Repository: okinawa-1962-nc3-near-launch — CC BY 4.0

Demonstrates that epistemic governance frameworks scale from AI systems to high-stakes historical research. Stewarding the testimony of my father, USAF missile technician John Charles Bordne, regarding the alleged 1962 Okinawa nuclear near-launch incident. Building public evidence architecture for historians using T1–T4 evidence tiers and transparent provenance discipline.

Status: Deployment Phase Initiated
Design Foundation: 70–75% code reuse from 3I/ATLAS epistemic infrastructure
Validation: Cross-platform portability verified (Claude, ChatGPT, Grok, Gemini)

Specialized AI systems inheriting unified epistemic governance and symbolic verifiability:

• Space Weather Sentinel — Real-time geomagnetic monitoring & forecasting
• NEO Tracker — Near-Earth object risk assessment & discovery
• Exoplanet Explorer — Habitable zone characterization & mission planning
• Dark Sky Finder — Light pollution mapping & observatory optimization
• Asteroseismology Analyst — Stellar oscillation analysis & internal structure
• Occultation Predictor — High-precision timing predictions
• Consciousness-Coherence Biofeedback — Phenomenological practice integration

Status: 🔒 Private (proprietary development)
Documentation: BordneSAI

Foundational logic for verifiably safe, self-aware, autonomously evolving LLM systems. Enforces perfect traceability, constitutional constraints, explainable reasoning, and verifiable safety.

Status: 🔒 Private (proprietary development)
Repository: BordneAI-EOS

The universal instruction set governing all BordneAI systems. Implements symbolic safety, stateless inference, and global alignment principles.

Status: 🔒 Private (proprietary benchmarking)
Repository: BordneSAI-Baseline-Testing

Proprietary benchmarking suite measuring LLMs along core optimization axes: self-scrutiny, symbolic coherence, multi-step integrity, exploit resistance.

Symbolic Verifiability
All inference paths must be explainable, auditable, and logically sound.

Immutable Trace Integrity
All internal decisions are recorded to an unalterable audit trail.

Ethical Autonomy by Construction
Systems enforce their own safety constraints; they cannot be coerced into violating core values.

Compliance-First Logic
Architected to satisfy and exceed global AI governance standards, including:

• EU AI Act
• NIST AI RMF

Different domains require different licensing:

• Proprietary (All Rights Reserved)

  • BordneSAI
  • EOS
  • Private core logic engines
  • Stateless symbolic runtime

• Creative Commons

  • 3I/ATLAS Gateway Guide: CC BY-NC-SA 4.0 (non-commercial research use)
  • NC3 historical research: CC BY 4.0 (open attribution, any use)
  • BordneAI Research Engine: CC BY 4.0 (open attribution, any use)

This separation maintains institutional clarity.
Every repository includes its own LICENSE file with explicit terms.

"All systems symbolic. All logic traceable."
"Verification is the first architecture."

© 2025 David Bordne.
All rights reserved unless otherwise specified.
Proprietary systems remain private; public research licensed per-repository.
See LICENSE.md for details.