Human.ExeAGI

Advanced Governance Intelligence

AGI R&D

The Human.Exe platform is built on a governed AI research corpus. Research informs the platform — it is not the product. Selected findings are published here as the platform matures.

Research Index · Jump To

Three tracks. Tap a card to jump straight to that catalogue.

ALIVE

Track A — Published

Public-release research. ACB benchmark + companion failure-variance study.

2 papers

BLIVE

Papers Corpus

Public + NDA-tier papers. Title-only previews until access is granted.

5 papers

CIN DEV

Signal Portfolio

38 active research threads across 12 domains. Detail access requires NDA.

12 domains · pre-NDA

Track A — Published

Research released for public review. No agreement required.

AGI-R-01v1.1

ALSI Inc. · AGI R&D Division · 2026

AGENTIC STUDY — FIRST RESULTS

Agent Cognitive Benchmark

Abstract

A standardised cognitive proficiency framework for AI code-generation agents. Structured test harness: deterministic scenarios, trap conditions, and scored delivery across two evaluation tracks. Track A results: top frontier models scored at ceiling. Track B (Agentic, April 2026): 14 models evaluated on autonomous delivery — first study to isolate delivery-surface behaviour as a primary outcome variable.

benchmarkcode generationagentic deliveryevaluation frameworkfrontier models

ALSI Inc. · human-exe.ca · Public ReleaseRead Paper →

AGI-R-01-S1v1.0Sub-study of AGI-R-01

ALSI Inc. · AGI R&D Division · April 2026

NEW

Agentic Interface Failure Variance

Abstract

A companion study to the ACB. 13 frontier models, agentic interface conditions, five observed failure strategies. Finding: the delivery interface — not model capability — was the primary determinant of outcome.

agentic interfacesfailure modesfrontier modelsdelivery surfaceinterface effects

ALSI Inc. · human-exe.ca · Public ReleaseRead Paper →

Track B — Papers Corpus

Selected research releases. Public papers are openly available; NDA-tier papers display title and series only until access is granted.

COG-HOW-AI-SYSTEMS-01

ALSI Inc. · AGI R&D Division

Public Release

Confident and Wrong

The Evaluation Problem in AI Cognition

Abstract

There is a property shared by almost every AI evaluation process in production today. It is not a flaw that any individual organization introduced. It was inherited — from how these systems were trained, what they were optimized for, and why that optimization produces a specific category of failure that standard review processes are structurally unable to detect.

ALSI Inc. · human-exe.caRead Paper →

GOV-GOVERNANCE-02

ALSI Inc. · AGI R&D Division

Public Release

The Collapse Signature

What Civilization History Teaches Us About AI Governance Failure

Abstract

In 2026, AI governance is typically framed as a recent problem requiring novel solutions. The framing is wrong. The problem is ancient. The solutions — successful ones — are documented across three thousand years of civilizational history. What is novel is the substrate. What is not novel is the structural failure mode.

ALSI Inc. · human-exe.caRead Paper →

MEM-MEMORY-PERSISTENCE-03

ALSI Inc. · AGI R&D Division

Public Release

Entropic Memory

Why AI Systems Forget What Matters First

Abstract

Something happens to AI-mediated production systems over time. It is not dramatic. It does not announce itself. It accumulates the way rust accumulates — invisible until it has already done its work.

ALSI Inc. · human-exe.caRead Paper →

COG-THE-BLANK-SLATE-04

ALSI Inc. · AGI R&D Division

Public Release

The Blank Slate Problem

What It Actually Means to Work With an Amnesiac System by Design

Abstract

Here is a situation that anyone who has worked extensively with AI systems will recognize immediately, even if they have never articulated it precisely.

ALSI Inc. · human-exe.caRead Paper →

COG-GOV-COHERENCY-IS-NOT-A-05

ALSI Inc. · AGI R&D Division

Public Release

Coherency Is Not a Feature

Why Consistent Output Is a Precondition for Trust, Not a Quality Level

Abstract

Consider two AI systems deployed in production.

ALSI Inc. · human-exe.caRead Paper →

Signal Portfolio — AGI R&D Division

38 active research threads across 12 domains. Detail access requires NDA.

Pre-NDA · Full portfolio on request

Non-Destructive · Infinite · Sustainable

HOW AI SYSTEMS LEARN AND REASON

Benchmarks, evaluation criteria, and architectural principles for AI cognition — grounded in governance-first design rather than benchmark gaming.

39 RQ

GOVERNANCE ARCHITECTURE FOR AI SYSTEMS

How rules, boundaries, and decision authority are encoded into AI system design at the architectural level — not as policy documents, but as engineering constraints.

34 RQ

MEMORY, PERSISTENCE, AND CONTINUITY

What principled, governed persistence looks like for AI systems — carrying forward context, identity, and prior state without compromising integrity.

38 RQ

IDENTITY, VERIFICATION, AND SECURITY

How to verify operator identity and confirm a system has not drifted from its authorized state — identity anchoring, verification protocols, and sovereign access control.

25 RQ

MULTI-AGENT SCALE AND COORDINATION

Governance, coherency, and trust across networks of coordinated AI agents — addressing emergent behaviours that single-agent frameworks cannot anticipate.

19 RQ

THE PHYSICS AND METAPHYSICS OF INTELLIGENCE

The mathematical and physical properties underlying cognition and emergence — and the structural boundaries of what artificial general intelligence can and cannot be.

52 RQ

BUSINESS FORMATION AND PLATFORM STRUCTURE

How a sovereign AI holding company structures itself for commercialization — formation architecture, licensing strategy, and the IP disclosure pipeline.

11 RQ

PLATFORM ENGINEERING AND DEPLOYMENT

The engineering path from research artifact to deployed system — architecture, infrastructure, integration, and sovereign hosting.

43 RQ

CROSS-THREAD COHERENCY

Internal consistency across 38 active research threads — ensuring no two frameworks contradict each other as the portfolio grows.

13 RQ

SPATIAL INTELLIGENCE

How AI systems operate in and reason about spatial environments — geographic and cartographic data, physical-world anchoring, and the convergence of intelligence and place.

62 RQ

SOVEREIGN MEDIA GENERATION

AI-generated media at production quality — video, audio, structured narrative. Generation architectures, quality governance, and distribution infrastructure independent of platform lock-in.

79 RQ

DIMENSIONAL SUBSTRATE FRAMEWORK

A formal mathematical and physical framework underlying the portfolio's governance architecture — dimensional geometry, modular arithmetic, and boundary condition theory applied as constitutional axioms. The 11th dimension is treated not as a compactified static axis but as a dynamic separating interval; the 12th as a generative boundary condition that encodes structural invariants across the system. Five working papers in progress targeting peer review.

24 RQ