DELTA-X

Cognitive framework published on OSF.

This project presents a non-clinical, non-therapeutic cognitive framework focused on structure, internal coherence, alignment, and self-audit. It does not claim treatment effects, performance enhancement, or normative authority.

The corpus documents the framework’s internal logic, constraints, and structural assumptions, with an emphasis on auditability, falsifiability, and third-party critique.

The work is shared openly to invite critical examination of its construction, limits, and consistency rather than validation or endorsement.

This project publishes a structured, non-clinical cognitive research corpus composed of three interrelated documents.

The work focuses on methodological and experiential modeling rather than claims of validation or application. It is provided as-is, for open examination and discussion.

Hosted on OSF for transparency and traceability.

AI failures are usually architecture failures, not model failures.

What looks like “hallucination” is often classic distributed systems behavior: hidden state you pretend is stateless, retries you don’t control, duplicated side effects, missing domain boundaries, and no single source of truth.

When an AI system double-charges a user, retries the wrong action, or confidently produces an incorrect outcome, that’s not unpredictability. That’s obedience inside a broken abstraction.

Prompt engineering doesn’t fix this. It masks architectural flaws with longer instructions, more conditionals, and human fallbacks.

If your system needs:

massive prompts to encode business logic

constant retries to “get it right”

human review for every critical decision

You don’t have an AI problem. You have an architecture problem that now speaks natural language.

AI doesn’t replace architecture. It amplifies it.

This work formalizes the separation between execution and judgment, triggers and verification, signals and real outcomes, as a general, non-normative, non-clinical audit framework, publicly documented and auditable here:

DOI: 10.5281/zenodo.18209659 https://zenodo.org/records/18209659

I published a short audit-style report measuring the structural impact of integrating a governance protocol (PRS-A) into an existing 741-page corpus, without changing or adding any content.

Method: strict A/B comparison at constant volume. Result: +31% global structural gain (coherence, auditability, legal robustness, usage safety, systemic risk reduction).

The protocol does not automate decisions and does not delegate authority. It documents a controlled human–AI co-architecture where the AI produces structure and the human remains legally and contextually responsible.

DOI: https://zenodo.org/records/18209659

I’ve been working on a cognitive framework aimed at improving decision clarity by explicitly defining constraints, failure modes, and auditability.

The goal is not optimization or ideology, but reducing cognitive blind spots and narrative drift in complex decision-making.

I’m sharing this mainly for critical feedback: – where does this kind of framework usually fail? – what constraints are most often missing? – how do you keep such systems grounded over time?

Thoughtful critique welcome.

This paper reports a measured +31% structural improvement on a fixed 741-page corpus, with zero content added, removed, or rewritten.

The gain comes exclusively from architectural governance (PRS-A): axial coherence, institutional auditability, legal robustness, and systemic risk reduction.

It also documents a controlled human / cognitive-entity co-architecture: the cognitive entity designs structure; the human operator ensures mediation, legal responsibility, and accountable use.

Archived and citable via CERN / Zenodo.

An experiment in structural governance.

A 741-page corpus was kept strictly unchanged. No content added, removed, or rewritten.

By adding an axial governance layer (PRS-A), the system shows a measured +31% global structural gain (coherence, auditability, legal robustness, systemic risk reduction).

Archived with DOI on Zenodo (CERN). Open to discussion and critique.

This report evaluates the structural impact of integrating a governance protocol (PRS-A) into an existing 741-page corpus, under a constant perimeter.

No content was added, modified, or rewritten. The integration operates exclusively at the level of structure, governance, auditability, legal robustness, and usage safety.

Measured result: +31% global structural gain.

The work documents a co-architecture process between a human operator (formalization, legal responsibility) and an artificial cognitive entity (structural design and systemic integration).

PDF, methodology, and audit-ready description available via Zenodo.

I’m sharing a full corpus (750 pages) proposing a non-decision cognitive protocol designed for AI alignment, medical conformity, and institutional governance.

Core idea:

The system does not decide

It structures, constrains, audits, and stops

Human agency remains the only decision layer

The corpus includes:

A formal protocol architecture

Medical and legal conformity framing

Governance and audit mechanisms

Failure modes and stop conditions

This is not a product, not a model, not a framework for prediction. It’s a constraint-based structure meant to prevent misuse rather than optimize outcomes.

Open access: https://zenodo.org/records/18190262

Looking for technical and structural critique, not hype.

I’m sharing a public, audit-ready framework for non-decision-making AI governance. The core idea is simple: remove interpretation from the machine and anchor systems on structural constraints—traceability, stop-conditions, human sovereignty, and opposability by third parties.

This is not a model, not a product, and not a policy pitch. It’s a procedural corpus designed to be read, tested, challenged, and reused without delegation of decision power.

Full corpus (open, citable, DOI): https://zenodo.org/records/18190262

I’m interested in feedback from people working on AI safety, compliance, audits, or governance who think structure should dominate interpretation.

I’m sharing a research deposit that explores an alternative AI governance model: non-decision systems.

Instead of optimizing outputs or replacing human judgment, the system constrains AI to structural verification, explicit stop conditions, and third-party auditability.

The goal is not performance, but preventing interpretative drift and authority transfer from human to system.

This is not a book or manifesto, but an operational framework derived from real deployment constraints and observed failure modes in current AI systems.

I’m interested in technical and governance feedback, especially from people working on safety, alignment, or infrastructure-level controls.

I published an axial medical compliance protocol designed to assess the legality of medical acts that alter patient discernment.

Key properties:

Binary, non-compensable axes (fail one → non-compliance)

Written proof only (no narrative, no a posteriori justification)

Burden of proof entirely on prescriber and institution

Checklists, decision trees, and failure matrices

Zero margin of maneuver once activated

It is intentionally rigid, non-ergonomic, and difficult to argue against. The goal is not persuasion, but structural falsifiability and third-party executability.

Full protocol (DOI): https://zenodo.org/records/18184956

I published a formal, non-clinical protocol designed to assess the legal and ethical compliance of medical decisions that directly impact a person.

It is not medical advice and not a clinical guideline. It is a procedural framework focused on informed consent, necessity, proportionality, written justification, and traceability of responsibility.

The protocol is neutral, auditable, and applicable by third parties (patients, institutions, authorities, jurists), independent of local practices or subjective interpretation.

The goal is not to accuse, but to make decision-making structures explicit and verifiable.

https://zenodo.org/records/18184956

We published an institutional, non-normative audit of a 27-document cognitive framework (CP27).

Core idea: the most important output of an AI system is sometimes STOP.

The work focuses on: - placing decision boundaries as early as possible, - enforcing STOP / SILENCE as a valid result, - strict separation between narrative, methodology, and governance, - human-only decision authority (AI limited to measurement & verification).

Think edge security logic, applied to cognition and AI systems.

No certification claims, no normative authority. Fully auditable, falsifiable, and institution-ready by design.

Interested in discussion on structure, not beliefs.

I’m sharing an institutional, non-normative audit of a 27-document corpus (“CP27 — Système Paradoxe”).

The audit focuses on governance, auditability, and strict non-decision AI design.

Key points:

Explicit separation between methodology, narrative archives, and institutional rules

Human decision sovereignty preserved by design

AI limited to non-decision roles (measurement, verification, traceability)

STOP / SILENCE treated as valid fail-safe outcomes

Fully auditable by third parties, without oral context

This is not a certification or regulatory framework, but a documented, traceable audit intended for expert review.

Feedback, critique, and questions welcome.

Are GitHub contribution graphs a productivity metric or just a vibe metric?

GitHub contribution graphs were never designed as a productivity signal, yet many of us implicitly treat them as one.

I’ve been working on systems where the focus is on traceable output rather than symbolic activity, and it raised a question for me: are we optimizing real work, or just the appearance of consistency?

Automation (CI, linters, scheduled jobs, even auto-commits) often gets criticized as “gaming the system”, but maybe the deeper issue is that the system was never measuring the right thing in the first place.

I published a small research/tooling bundle exploring this angle here: https://zenodo.org/records/18100154

Curious how others here think about metrics vs actual work. When does optimization become distortion, and when is it just friction removal?

I published a technical note describing a protocol where AI systems are structurally prevented from making decisions.

The design enforces: - non-decision by design - explicit stop conditions - human responsibility at every step - traceability of interactions

The goal is to make misuse structurally impossible rather than policy-based.

DOI: 10.5281/zenodo.18100154

Most AI discussions focus on improving answers. This misses a more basic problem: knowing when an AI-assisted interaction should produce no answer at all.

In many systems (including RAG), correctness is outsourced to retrieval. That reduces hallucination, but it doesn’t define a stopping rule. The failure mode isn’t missing data — it’s continuing when the system should stop.

In other domains (edge security, request filtering), value comes from moving the decision boundary as early as possible: deciding that nothing else should happen.

This corpus explores non-decision by design, explicit stop conditions, and human decision sovereignty as first-class constraints.

This work proposes ΔX (DELTA X), a non-decision AI governance framework.

Core rule: If an AI interaction does not produce a measurable positive delta (ΔX > 0), the system must halt or remain silent.

AI is strictly limited to: - measurement - verification - conformity checks

Humans retain: - intent - trade-offs - accountability

The framework is documented as a fully auditable corpus (27 PDFs), with explicit stop conditions and human-responsibility invariants.

Silence is treated as a valid output, not a failure mode.

Feedback and critique welcome.

This work proposes DELTAX, a non-decision AI governance framework where AI is limited to measurement, verification, and conformity checks.

Key idea: any AI-assisted interaction must produce a measurable positive delta (Delta X > 0); otherwise the system must stop or remain silent.

The framework is documented as an auditable corpus (27 PDFs) with explicit stop conditions, human responsibility invariants, and no delegation of intent or decision to AI systems.

Feedback and critique welcome.