Research programme

Five papers, five manifestos

A systematic research programme on governed intelligence for regulated industries. All papers co-authored by Witold Reichhart and Arnaud Gelas.

The causal spine

Enterprise AI fails because of dynamics blindness (A) → the resolution is architectural (B) → the architecture works because ten traditions converge on requirements (C) → the practitioner methodology includes epistemic immunity (D) → when the architecture runs at sufficient depth, it produces governed initiative (E).

SSRN working papers

Paper A — SSRN 2026

Dynamics Blindness: When AI Is Locally Correct and Globally Non-Compliant

Diagnoses the architectural failure mechanism in enterprise AI. LLMs process tokens without tracing causal chains through organisational dependencies. Chain-of-thought, RAG, tool use, and multi-agent systems do not add the missing causal infrastructure. The problem is structural, not parametric.

Reichhart, W. & Gelas, A. (2026)

Read on SSRN →

Paper B — SSRN 2026

The Predictive Organization: Architecture for Enterprise Intelligence

Specifies the architectural resolution to dynamics blindness. A tripartite structure - Map, Physics, Player - coupling neural perception with symbolic reasoning, operating on claims-based knowledge with prevalence weighting.

Gelas, A. & Reichhart, W. (2026)

Read on SSRN →

Paper C — SSRN 2026

Build the Medium: Why Organizational Intelligence Is Mechanism, Not Metaphor

Theoretical foundations. Ten independent traditions - from cell biology to social systems theory - converge on the same architectural requirements for organisational intelligence. Introduces the capability/fertility distinction and the autonomy-to-initiative transition.

Reichhart, W. & Gelas, A. (2026)

Read on SSRN →

Paper D — SSRN 2026

Governed Intelligence Architecture for Institutional AI

Practitioner methodology. The Governed Intelligence Lifecycle - Ingest, Consolidate, Curate, Expand, Apply - with an epistemic immunity framework protecting against six systemic knowledge failures. Introduces epistemic operational risk as a distinct risk category.

Gelas, A. & Reichhart, W. (2026)

Read on SSRN →

Paper E — SSRN 2026

From Autonomy to Initiative: The Ultimate Prize in Agentic Engineering?

Capstone paper distinguishing autonomy (independence in execution) from initiative (perception of what matters through immersion). Defines three formal conditions for governed initiative: fertile pattern density, constraint-legible action space, progressive governance relocation.

Reichhart, W. & Gelas, A. (2026)

Read on SSRN →

The Agentic Governance Stack

Five public manifestos

A five-layer governance framework spanning engineering practice through enterprise transformation. Each layer has a published manifesto with principles, values, and implementation guidance.

Layer 1

Agentic Engineering Manifesto

Principles for building systems where humans steer intent, agents execute within governed boundaries, and verified outcomes are the only measure.

Gelas, A.

Layer 2

Agent Software Development Lifecycle

The ASDLC - development lifecycle for agent-based systems in regulated environments.

Gelas, A.

Layer 3

Agent Product Lifecycle

The APLC - product governance from qualification through deployment, monitoring, and revalidation.

Gelas, A.

Layer 4

Intelligence Governance Manifesto

Governed intelligence as an operational discipline. Six values: governed claims over documents, traceable provenance over trusted sources, preserved contradictions over forced consensus.

Reichhart, W. & Gelas, A.

Layer 5

Agentic Enterprise Manifesto

Enterprise-level transformation principles for organisations deploying AI agents at scale across regulated operations.

Reichhart, W. & Gelas, A.

Key concepts

Named contributions

Original concepts introduced across the research programme.

Dynamics Blindness

Governed Intelligence Lifecycle

Map / Physics / Player

Epistemic Immunity

Capability / Fertility

Epistemic Operational Risk

Autonomy-to-Initiative

Governance Relocation

Living Medium

Domain Graph