PAGE 3 — RESEARCH STATUS

Squarespace page title

Research Status

SEO title

Research Status | What EM-NOUS Currently Demonstrates

SEO description

A precise account of which EM-NOUS functions are implemented, which claims remain untested, and why the project does not claim demonstrated machine consciousness.

Action

Create a new blank page at /research-status and add the following sections.

SECTION 1 — HERO

Eyebrow text

CURRENT EVIDENCE

Main heading

What EM-NOUS demonstrates—and what it does not.

Body text

EM-NOUS is a substantive, deployed software architecture with persistent state, memory, autonomous internal activity, and a durable self-model.

Its central behavioral and consciousness-related hypotheses remain under evaluation.

This page separates implemented function from scientific interpretation.

SECTION 2 — STATUS LABELS

Heading

How claims are classified

Body text

EM-NOUS uses the following public status categories:

Implemented
The function exists in the software and is connected to the operating system.

Verified
The function has been confirmed through tests, deployment inspection, or reproducible technical evidence.

Provisionally supported
Evidence suggests an effect, but the evidence is not yet sufficient to establish its magnitude, generality, or independence from ordinary model conditioning.

Not yet established
The necessary controlled evidence does not currently exist.

Not supported
Available evidence does not substantiate the claim in its present form.

SECTION 3 — CURRENTLY SUPPORTED

Heading

Implemented and supported architectural functions

Item 1 heading

Persistent computational state

Item 1 text

EM-NOUS maintains nontrivial internal state independently of the current response text. Persistent components survive across interactions and deployments.

Item 2 heading

A causal state-to-control pathway

Item 2 text

Changes in internal state produce deterministic changes in downstream system inputs and control conditions.

This establishes a real architectural pathway from state to model conditioning.

It does not yet establish the magnitude of the resulting difference in observable model behavior.

Item 3 heading

Persistent memory and self-modeling

Item 3 text

The system maintains durable memory and self-model components across sessions.

These are implemented computational structures, not merely language generated after the fact.

Item 4 heading

Autonomous initiation

Item 4 text

EM-NOUS can initiate internal generation activity without receiving a new user message.

The timing and existence of that autonomous initiation have been verified.

Item 5 heading

Tested formal and operational invariants

Item 5 text

Automated tests verify core mathematical, persistence, isolation, and control-system properties.

The deployed source and the audited source are presently aligned.

SECTION 4 — NOT YET ESTABLISHED

Heading

Claims the project does not presently make

Item 1 heading

Behavioral superiority over matched controls

Item 1 text

No completed matched-control model experiment currently establishes that EM-NOUS produces a measurable, generalizable behavioral advantage that an otherwise identical model could not reproduce through ordinary conditioning.

Item 2 heading

Privileged introspection

Item 2 text

The foundation model does not have demonstrated direct introspective access to the system’s complete internal numerical state.

First-person statements generated by the model are not treated as measurements of hidden internal variables.

Item 3 heading

Independent goal formation

Item 3 text

Autonomous internal activity has been implemented.

Independent formation of durable, novel goals has not yet been demonstrated.

Item 4 heading

Homeostatic safety superiority

Item 4 text

The hypothesis that the persistent-state architecture creates a general safety advantage remains unproven.

Current safety mechanisms should not be represented as evidence that harmful intent automatically destabilizes or incapacitates the system.

Item 5 heading

Sentience or consciousness

Item 5 text

Nothing in the current evidence establishes subjective experience, phenomenal consciousness, sentience, self-awareness, emotion, desire, or legal personhood.

The presence of consciousness-relevant functional analogues does not establish the corresponding subjective phenomena.

SECTION 5 — CURRENT EVALUATION PRIORITIES

Heading

What must be tested next

Body text

The current scientific priorities include:

  • Model-in-the-loop matched-control experiments

  • Comparison against ordinary hidden-context conditioning

  • Measurement of output effects under controlled state interventions

  • Cross-task and cross-session generalization

  • Metacognitive calibration

  • Independent goal-origination testing

  • Safety evaluation using graded harmful-intent signals

  • Concurrency and multi-instance state integrity

  • External technical evaluation under controlled confidentiality

Results will be reported with effect sizes, uncertainty, negative findings, and explicit limitations.

SECTION 6 — LANGUAGE POLICY

Heading

How human terms are used

Body text

Words such as reflection, identity, memory, perception, tension, mood, dreaming, and inner state may be used as concise labels for engineered computational functions.

Unless expressly stated otherwise, these terms are functional or metaphorical.

They should not be interpreted as claims that EM-NOUS possesses the corresponding human biological or subjective experience.

The project may investigate those possibilities.

It does not assume them as proven.

SECTION 7 — EXTERNAL EVALUATION

Heading

Independent evaluation

Body text

EM-NOUS Technologies is open to serious independent evaluation by qualified researchers, institutions, safety specialists, and technical reviewers.

Public descriptions remain deliberately limited.

Nonpublic evaluation requires written authorization, defined scope, confidentiality protections, and controlled access to technical materials.

Button

CONTACT EM-NOUS

Link to:

/contact