Organizational Sentience

ABSTRACT

WP 0 established the integration condition. This paper takes the first of the seven individual conditions: being. The thesis: an organization that cannot register and report its own state cannot correct it either — however capable its parts.

The paper draws on four AGI-26 contributions that hit the same point from different directions: metacognition is a component of the architecture and not a by-product of capability (Laird, Lebiere, Rosenbloom & Stocco); self-models can be measured via proxy quantities (Bergmann; Salvi & Wolfson; Samsonovich & Khabarov); and a self-modifying system converges on a mathematically determined boundary whose two directions of failure are opposites (Tóth-Pócs). From this follows Condition 1 and a test that is cheap to run and hard to pass.

Keywords: Organizational Sentience · metacognition · self-model · Coverage Ratio · cognitive closure · LEAD21 · leadership climate

Suggested citation: Huesch, G. (2026). Organizational Sentience. Working Paper Series 2026, Series 1 "The Superintelligent Organization", WP 1. Global Leadership Consultancy.

PART 1 · THE FINDING

1 THE QUESTION

You drive forty minutes of motorway. You hold your distance, change lanes, respond to a brake light three cars ahead. And as you get out you cannot say what the route was like.

Nothing about this is a mistake. The drive was safe, the responses correct, the destination reached. What is missing is something else: access to your own state during the drive.

Many good companies drive that way. They hit targets, correct deviations, survive recessions. Their reflexes are excellent. What is missing is the ability to say how the market currently appears to them — and to notice when that appearance has gone stale.

The question of this paper: how can one tell that an organization perceives itself — and does not merely react?

2 WHAT AGI-26 SHOWS

Metacognition is a component in its own right, not a by-product. Laird, Lebiere, Rosenbloom and Stocco — the originators of the Common Model of Cognition — treat metacognition as a component of the architecture. That is a decision against the obvious assumption that a sufficiently capable system brings self-report along by itself. It does not. It has to be built.

For organizations this means: more data, more layers, more dashboards do not produce self-perception. They produce more reflexes.

Self-models can be measured — via proxy quantities. Three contributions work on this. Frank Bergmann proposes a proxy metric based on self-models. Salvi and Wolfson search for machine correlates. Samsonovich and Khabarov build a multi-agent model of self-awareness on a language-model basis. The common move: no one tries to measure consciousness directly. Everyone measures whether a system holds a model of itself and whether that model influences its behavior.

That is exactly the grip an organization needs. The question is not whether a firm "feels". The question is whether it holds a model of its own state that changes what it does.

Self-access has two opposite failure modes. With HORISMOS, Gáspár Tóth-Pócs develops a diagnostic quantity that is immediately usable for leadership: the Coverage Ratio ρ. It measures how completely a system believes it has mapped its own possibility space.

The finding that counts: health lies strictly between the extremes. A system that holds its space to be fully mapped is not mature — it is closed. And Tóth-Pócs adds a sentence that explains the whole question of routines:

"Learning is expensive; forgetting is cheap."

Distance in information space is directed. Ascent costs, descent is free. That is why knowledge not held in a routine decays on its own — without anyone doing anything wrong.

Translated: the unit of your company that most surely knows how its market works is the candidate for ρ → 1. That is not maturity. That is cognitive closure.

3 WHAT ORGANIZATIONAL RESEARCH HAS SAID FOR DECADES

Weick and Roberts (1993) show, using aircraft carriers, that collective capacity for judgment arises from the way actions are related to one another — heedful interrelating — and not from the brilliance of individuals. Argyris and Schön (1978) separate first-order from second-order learning: in the first case you correct the deviation, in the second the theory that produced it. Only the second presupposes self-access. Walsh and Ungson (1991) describe where organizational memory can reside at all — in people, culture, procedures, structures, the work setting and external archives. Their uncomfortable insight: five of these six stores are silent. They give an account only when someone asks them.

4 THE SIX PROPERTIES

From both literatures, six properties can be derived, each with a failure pattern by which it is recognized.

5 THE TEST

Does your organization lose its capacity for judgment when a central leader departs?

Cheap to run, hard to pass. And it has an uncomfortable property: one usually knows the answer before asking.

If the answer is yes, the capability never resided in the organization. It resided in a head, and the organization grew used to that head being present. This is no reproach to the person — on the contrary, it is often the consequence of someone having been very good. But it is a statement about the architecture.

The more precise test distinguishes four stages:

1. The organization does not notice the deviation.

2. It notices, and a person corrects.

3. It notices, and a routine corrects.

4. It notices that its way of noticing has gone stale.

The jump from 3 to 4 is what this series is about. It is also the only one not reached by greater effort.

6 ONE EXAMPLE, WORKED THROUGH

Illustrative example, not a case report.

A machine builder has run the same monthly reporting for eleven years. The figures are clean, the meeting takes ninety minutes, deviations are explained and remedied. Stage 3, cleanly achieved.

Then the market shifts: customers no longer buy machines, they buy uptime. The reporting does not notice — not because the figures are wrong, but because it counts machines. The category "unit sold" is the lens through which the organization looks, and the lens is not on the agenda.

Eighteen months later, sales says it clearly for the first time. The question is not why it took eighteen months. The question is which body would have been responsible for reviewing the organization's own way of measuring. There was none.

ρ was close to 1. The organization held its space to be fully mapped.

7 HOLD ON

Objection 1: Isn't this just good reporting?

A reporting system captures what it was built to capture. Self-perception means being able to question the manner of capture itself. No dashboard reports that it is measuring the wrong quantity.

The advantage: whoever once has a routine in which the manner of measuring is reviewed needs that routine less often than they fear. Twice a year is enough — but it must exist, or it never happens.

Objection 2: This sounds like navel-gazing. We have a business to run.

The objection is right if self-perception were meant as a permanent state. It is not. It is a beat. The organization runs on reflexes ninety-nine per cent of the time — rightly so, anything else would be slow. The one per cent decides whether the reflexes still match the terrain.

The advantage: that one per cent is cheap. It costs a meeting, not a programme.

8 THE CONDITION

Condition 1 · Self-model An organization can correct its state only as far as it can register and report it. Capacity for judgment bound to a person is not a property of the organization but a loan.

PART 2 · WHAT LEAD21 ANSWERS

CRITICAL APPRAISAL

Five of the six properties have a named instrument in LEAD21. The sixth has none, and that is the most interesting point in this paper.

Openness — the capacity to hold one's own possibility space as incompletely mapped — is not built into LEAD21 as an instrument. The system brings problems quickly to the right place. It has no mechanism that asks whether the places are still the right ones.

To this comes the familiar limitation: the mapping comes from the developer of the system. And the survey on which the "self-perception" row rests is an expectation survey with self-report, not a measurement — 143 leaders across nine organizations, 18 dimensions, running since 2016. Exactly one survey in LEAD21 is measured, and it belongs to WP 4.

WHAT THE SERIES GIVES BACK TO LEAD21

Two concrete additions.

First, an instrument for openness. A fixed routine, twice a year, in which not the figures but the manner of measuring is reviewed: which category do we count, since when, and what do we thereby fail to count? Proposed metric: the number of measures changed in the last three years. If it is zero, ρ is close to 1.

Second, a survey of persistence. How long do the mechanisms carry after the programme ends? This has never been surveyed. Without that figure, "mission continuity" is substantiated for the duration and a conjecture beyond it.

FALSIFIABLE PREDICTION

Prediction 1. Organizations with a fixed routine for reviewing their own manner of measuring change more measures over the observation period than comparable organizations without it — and notice market shifts earlier, measured by the interval between first internal mention and formal adoption into planning.

Falsifiable: if no difference appears, openness is either not producible by a routine or not causal for early detection.

PART 3 · ABOUT GERALD HUESCH

Whoever claims that an organization can register its own state must show with what. Gerald Huesch has surveyed leadership climate since 2016 across 18 dimensions; 143 leaders from nine organizations have taken part so far. This is an expectation survey with self-report, not a measurement — the distinction is stated explicitly, because it is what separates an instrument from a promise. Six core dimensions for future measurement have emerged from it.

The order of magnitude at stake is stated on gerald-huesch.com: up to 50 billion leadership touchpoints every working day, up to 12.5 trillion a year, across roughly 500 million leaders worldwide and 25 to 50 leadership moments per leader per day. The largest human coordination task in the world — and largely unsystematized.

Systems shape behaviour. Through the selection of the right instruments, mechanics and leadership codes — derived from values and leadership guidelines — the behaviour of everyone involved is synchronized and aligned. Not through motivation, but through a system that produces the right behaviour when applied. Sales has Salesforce. Finance has SAP. Leadership has LEAD21.

About the author. Gerald Huesch is the creator of the leadership operating system category – "one of the most professional and creative minds in the domain of leadership and difficult change projects" (Boston Business School). With LEAD21 he built the first operating system for leadership – 21 evidence-based methods, the comprehensive solution and the missing link for complete transformation, extreme growth and high-volume post-merger integration: a unique leadership system for human coordination at scale, in which alignment and culture emerge as a by-product. He has trained more than 30,000 executives in over 30 countries (NASA, Rolls-Royce, Airbus, Omnicom and others) and delivers executive leadership training, leadership development and executive mentoring for executives in Germany, the Americas, Asia Pacific and the UAE. Participants from 300+ companies report +87% clarity, +58% efficiency, +50% performance, +46% alignment and −50% negative stress after LEAD21 (participant-reported); a one-year study with 567 leaders (Prof. Beate Hanson & Prof. Mohit Bhandari, McMaster University) showed improvement in 8 of 9 climate dimensions. Dr. Charles Pellerin (NASA) on LEAD21: "Unique in the world, and a capability that nobody else has." gerald-huesch.com/about-en

PART 4 · OUTLOOK

STATUS OF THE SERIES

WHAT COMES NEXT

WP 2 · Perspective Before Data. Modes of Disclosure — not what is measured, but how a situation is given. This paper showed that an organization must be able to review its manner of measuring. The next shows why the same figure carries two incompatible decisions without either party being wrong.

OPEN QUESTIONS FROM THIS PAPER

1. Is the number of changed measures a usable proxy for ρ, or does it only measure restlessness?

2. Are there organizations that reach stage 4 without a routine of their own — because a market forces them often enough?

3. Can "memory" be separated in practice from "memory of reasons" without inflating the minute-taking?

NOTES

On method. Conception, argument and responsibility rest with the author. An AI system (Claude, Anthropic) was used as a working instrument for research, structuring, the transposition of the AGI-26 contributions and the bilingual drafting, with the explicit instruction to assess the relation of LEAD21 to AGI critically and with the greatest possible objectivity, including against the author's interest. Readers are expressly invited to evaluate the contribution of LEAD21 to AGI critically, with maximum objectivity of their own, and to draw their own conclusions.

Declaration of interest. The author is the developer and owner of LEAD21 and has a commercial interest in its dissemination. Every statement about LEAD21 is assigned a degree of coverage and an evidence status.

On sources. The author evaluated the posters he documented at AGI-26 as well as titles, abstracts and the proceedings tables of contents — not the full texts. Verbatim quotations come from the posters. The transfer to organizations is an analogy by the author. Where a proceedings entry could not be verified first-hand, this is noted.

REFERENCES

AGI-26

Bergmann, F. W. (2026). Functional consciousness: A proxy metric using self-models. In M. Iklé, A. Franz, A. Kemendo & H. Lowy (Eds.), Artificial General Intelligence. AGI 2026, Part I (LNCS 16854, pp. 79–95). Springer.

Bringsjord, S., et al. (2026). Toward detecting and measuring consciousness in service of formal theories of intelligence. In Iklé et al. (Eds.), AGI 2026, Part I (LNCS 16854, pp. 96–115). Springer.

Laird, J., Lebiere, C., Rosenbloom, P., & Stocco, A. (2026). Unified, comprehensive metacognition within the Common Model of Cognition. In Iklé et al. (Eds.), AGI 2026, Part II (LNCS 16855, pp. 1– 15). Springer.

Salvi, R. E., & Wolfson, O. (2026). Discovering machine correlates of consciousness. In Iklé et al. (Eds.), AGI 2026, Part II (LNCS 16855, pp. 237–255). Springer.

Samsonovich, A. V., & Khabarov, D. L. (2026). BICA-inspired multiagent model of self-awareness based on LLM. In Iklé et al. (Eds.), AGI 2026, Part II (LNCS 16855, pp. 256–276). Springer.

Tóth-Pócs, G. (2026). HORISMOS: Self-representation and the derived constitutional boundary in enriched cognitive systems. Conference poster, AGI-26, San Francisco State University, 27–30 July 2026. — Poster documented first-hand; proceedings entry open.

Organizational and leadership research

Argyris, C., & Schön, D. A. (1978). Organizational learning: A theory of action perspective. Addison-Wesley.

Walsh, J. P., & Ungson, G. R. (1991). Organizational memory. Academy of Management Review, 16(1), 57–91.

Weick, K. E., & Roberts, K. H. (1993). Collective mind in organizations: Heedful interrelating on flight decks. Administrative Science Quarterly, 38(3), 357–381.