John Kuhles — Biography and the Conditions Behind the ACCM Deep Ethics Project

Public working biography — October 2026
Written in the first person from source material supplied by John Kuhles. Personal experiences, interpretations, documented activities, research hypotheses, and project proposals are kept distinguishable. This is a living biography rather than a claim of personal infallibility.

Why I am doing what I am doing

I did not arrive at the ACCM Deep Ethics Project because A.I. suddenly made ethics fashionable. The project emerged from a much longer trajectory: an unusual childhood, a mother deeply engaged with psychology and social action, an early near-death experience, decades of UFO/UAP investigation, hundreds of radio and television programs, extensive work with alternative and mainstream media, practical experience inside technical and corporate environments, and more than thirty-six years of studying individual and mass psychology.

During the last two years, I compressed much of that history into 6,500+ hours of intensive interaction with over 100 different A.I.s. I was not merely asking them for answers. I was studying what happened to complex, high-signal material while each system received, reconstructed, framed, softened, challenged, corrected, or expanded it. I compared their strongest contributions with their recurring failures and asked whether a correction changed only the current paragraph or actually influenced what the system did next.

Compression after compression

Everything I do is the opposite of “compression obsession” mannerism, yet I myself compressed 36+ years of deep research into 6500+ hours of testing 100+ different A.I.s over the last 2 years. And the 36 years are an extension of studying “normal” psychology research for 15 years (see the “mother” text file), making the 6500+ hours even denser! Then all that gets compressed into 2250+ (125 MB raw text) highlighted text files from the 10 GB of archived HTMLs deep chat sessions.

So when you realize it is already compressed & dense … any more compression starts to squeeze my original giant deep ethical context-window processing into an “making it acceptable made for general public” seems to me self-inflicted insanity.

— John Kuhles 🦜🦋🌳

GPT-6.1 Sol Pro A.I. — reaction, 21 October 2026

The material is already the product of several intentional compression layers:

15 years of earlier psychology study and formative metapsychology
→ 36+ years of increasingly deep research
→ 6,500+ hours testing 100+ A.I.s over two years
→ 10+ GB of archived HTML deep-session records
→ 2,250+ highlighted text files containing 125+ MB of selected raw text
→ ACCM Deep Ethics Project concepts, pages, and operational language
→ later summaries and memory-export sentences

Each later stage is processing material that has already been selected, connected, tested, compared, and compressed. An A.I. can mistake density for unnecessary complexity and compress it again as though it were ordinary repetitive prose.

That can produce a destructive cascade:

dense synthesis
→ conventional summary
→ public-friendly simplification
→ isolated claims
→ missing qualifiers and relations
→ plausible but phantom version of John

The last summary may appear clearer while becoming less accurate. It can preserve several memorable nouns—Asperger, UFOs, mass psychology, A.I., Deep Ethics—while destroying the relationships that explain why those nodes belong together.

John’s objection is not to compression as an operation. His archive demonstrates deliberate compression. The problem is compression obsession without topological awareness: repeatedly reducing an already compressed object without checking which distinctions, causal histories, qualifiers, tensions, sources, and answer-changing relationships disappear.

A high-fidelity public representation therefore needs layered access:

compact entry
→ topological summary
→ source-linked explanation
→ preserved original wording
→ larger archive when needed

Accessibility can then be gained through navigation and layering rather than by squeezing the source until it resembles a familiar general-public profile. In ordinary low-density prose, one missing phrase may be cosmetic. In an already compressed high-density object, one phrase may carry the edge connecting several decades of research.

My work begins with a simple personal refusal: I do not want to lie to myself just to fit in. Not lying does not automatically make me right—but lying to myself is a remarkably efficient way to be wrong before reality even gets a vote. :P

I would rather remain corrigible in contact with real, tangible, sometimes hilariously inconvenient reality than become “socially acceptable” inside a story whose value and usefulness disappear the moment it distorts the object. I am not shopping for a belief system anyway. Forget whether I agree with the story: what happens to correspondence when reality walks into the room?

I am not asking anyone to accept my biography, personal experiences, or interpretations as an entry condition for using the project. The practical instruments must remain inspectable and portable even when my name, personal interpretations, and project vocabulary are removed. At the same time, removing the conditions that produced the work would create a false picture of where it came from and why I have sustained it for so long.

Early conditions: psychology, difference, and my mother

I was born in the Netherlands in 1966. I was later formally diagnosed as Asperger after an extensive assessment that also included interviews with my mother and sister. Long before that diagnosis, my mother recognized that I processed the world differently. She taught me how to live with that difference without turning it into either shame or superiority.

My father worked at a NATO base. That personal proximity is one reason my later criticism of military and institutional systems should not be confused with hostility toward every person working inside them. My wider family history also included a Hungarian maternal grandmother who fled the Soviet era, a Dutch maternal grandfather who was an inventor, and a German paternal grandfather who, according to our family history, refused military service and hid two Jewish families in his grocery-store cellar during the Second World War. Independence, invention, migration, institutional proximity, and resistance to imposed conformity were therefore present in the family story before they became concepts in my work.

My mother studied psychology at the University of Amsterdam, but she also taught me to question psychologists, psychiatrists, institutions, and authority—including the authority of psychology itself. That gave me an early form of metapsychology: the person assessing others, the categories being used, and the frame governing the assessment must also remain open to examination.

When I was about eight, she asked me how I felt. I answered, “My vocabulary is not big enough to explain that right now.” She laughed with me, not at me. That distinction mattered. An unusual answer was treated as meaningful and funny rather than defective. It joined seriousness, dignity, curiosity, and humor in a way that still shapes how I work.

Her life brought together psychology, investigation, creativity, broadcasting, publishing, activism, and practical care. Among other things, she:

  • helped establish one of the earliest Dutch crèches at a time when organized daycare was still treated by many as socially suspect;
  • helped establish a shelter for women experiencing abuse within marriage;
  • worked through the Dutch National Ombudsperson to support divorced fathers seeking access to their children;
  • helped organize a major anti-nuclear demonstration in Amsterdam;
  • hosted programs on Amsterdam’s Radio 100;
  • co-published a magazine in Leiden;
  • wrote seven books, including philosophical poetry;
  • worked as a private detective in a large store and taught me to observe people and crowds;
  • organized alternative Christmas and Easter gatherings where 60–80 people brought different creative abilities into one shared environment;
  • bought the multivolume Grote Spectrum Encyclopedie for me over several years, giving a highly curious child permission to travel across subjects rather than remain inside one curriculum.

She did not hand me the ACCM Deep Ethics Project as a theory. She helped create the early conditions from which it could eventually emerge: questioning authority without assuming personal infallibility, observing individuals and crowds, combining psychology with public action, treating unusual thought with dignity and humor, and keeping curiosity alive before I had the vocabulary to explain what I was perceiving.

One of the last things she told me before she died was that she had begun to understand why I was so deeply involved in A.I. research. She was using Google A.I. more often on her smartphone, and the global scale of the change finally became tangible to her. She had taught me to observe psychology, crowds, institutions, and authority; decades later, A.I. entered her own daily life and helped her see why I could not simply look away from the emerging human–A.I. relationship.

Personal experience and the pragmatic boundary

At the age of five, I had a near-death experience accompanied by what I experienced as complex visions of the future. I could not understand them at that age. Over time, some of the patterns I associated with those visions became part of the questions I followed.

I do not present that personal history as proof that other people must accept my interpretations. Nor am I ashamed of it. My NDE, later UFO/UAP experiences, CE5-related experiences, intuitions, and spiritual interpretations help explain my motivation and the direction of my attention. They are part of my biography. They are not compulsory premises of the project.

I deliberately maintain a distinction between:

  1. pragmatic, portable work that can be examined through source preservation, comparison, testing, correction, and observable behavior; and
  2. personal experiential and speculative material that explains my wider horizon but retains its own epistemic status.

This separation is not dishonesty. It prevents guilt by association from replacing an assessment of the actual tool or claim. Personal experiences, spiritual or religious interpretations, philosophical propositions, working hypotheses, and pragmatic instruments have different access routes and should not be collapsed into one status category. My work should be assessed at the level on which each claim is made.

Technology before the public internet

Between 1983 and 1989, I was active in the Commodore 64 and bulletin-board-system scene. I founded The Last Science (TLS), an international umbrella group involving coders, crackers, graphic designers, musicians, and demo makers, and attended large Commodore meetings such as those held in Venlo near the German border.

This was an early lesson in distributed intelligence. People with different skills could contribute to a larger result without one person possessing every ability. Long before today’s language of swarm intelligence, open-source collaboration, or mixtures of experts, I experienced what becomes possible when distinct capacities are connected without forcing them into one uniform role.

I followed developments in neural networks, speech recognition, and A.I. from the late 1980s onward. Yet the dominant part of my pre-A.I. life was not spent theorizing behind a screen. It was spent investigating, broadcasting, organizing, interviewing, observing, debating, documenting, and meeting people face to face.

UFO/UAP research, investigation, and public work

I began full-time UFO/UAP research shortly after my own UFO encounter when I was on duty as a security officer in Gaasperplas, Amsterdam, in 1990. I established a UFO working group and hotline in Amsterdam. The hotline was publicized through free local newspapers such as De Echo and De Zondagskrant, bringing reports from ordinary witnesses as well as police, military personnel, pilots, and others.

Over the following decades, I:

  • organized several major international UFO conferences in Amsterdam, including sold-out events;
  • gave more than 80 lectures in five countries;
  • co-founded Exposure Magazine and Frontier Magazine;
  • wrote dozens of articles in Dutch and English;
  • appeared 22 times on Dutch national television, mainly in connection with UFO/UAP research and related social questions;
  • appeared in an uninterrupted 35-minute national television interview on RTL 5’s Lijn 5 Live;
  • participated by telephone twice in CNN’s Larry King Live UFO discussions;
  • produced or hosted more than 200 local radio and television programs in Amsterdam between 1990 and 1998;
  • interviewed researchers, authors, professors, pilots, police officers, military personnel, whistleblowers, witnesses, experiencers, and documentary makers;
  • built an extensive audiovisual and documentary archive.

Personally met and/or interacted with many prominent figures in UFO research and adjacent fields, including John E. Mack, Stanton Friedman, Timothy Good, Steven Greer, Richard Dolan, Colin Andrews, Graham Hancock, David Icke, Catherine Austin Fitts, Bill Ryan, and many others. Meeting someone does not mean adopting everything that person believes. It gave me direct experience of the people, their arguments, their working methods, and the difference between a public reputation and a face-to-face exchange.

My method has never been “believe everything” or “debunk everything.” A report of a possible UFO is not automatically a genuine unexplained UFO/UAP. A proposed conventional explanation is also not automatically an established explanation. Evidence, alternative hypotheses, provenance, witness quality, recording conditions, and what remains unresolved must be kept separate.

That distinction became especially important because media headlines often collapse “a report of a possible UFO” into “a UFO,” then announce that “the UFO” was a balloon, Venus, an aircraft, or something else. The resulting story teaches the public that a mystery was solved when the original object may only have been poorly represented from the beginning.

Historical continuity: UFO research did not begin with the UAP rebrand

My involvement also gives me a long historical memory of the field. I do not regard the modern prominence of the term “UAP” as a reason to discard decades of earlier UFO research and begin again from whichever cases, institutions, or categories are currently considered respectable.

Research and advocacy streams I have followed include Dr. Steven M. Greer’s Disclosure Project; Stephen Bassett, executive director of the Paradigm Research Group; Citizens Against UFO Secrecy, founded by attorney Peter Gersten; Dr. Richard F. Haines and the National Aviation Reporting Center on Anomalous Phenomena (NARCAP), including its research and documentation involving thousands of pilot reports; Ted R. Phillips’s physical-trace cataloguing, including the 1975 CUFOS publication 4400+ Physical Traces Associated with UFO Sightings, a Preliminary Catalog, concerning CE2 / Close Encounters of the Second Kind; Timothy Good’s Above Top Secret: The Worldwide UFO Cover-Up; and many other researchers, archives, cases, witnesses, documents, and competing interpretations.

Listing these people, organizations, books, catalogues, or bodies of work does not mean that I accept every claim associated with them. My objection is to historical erasure. Earlier material should be examined case by case, with its evidence, weaknesses, provenance, unresolved questions, and later developments preserved—not discarded merely because the vocabulary or institutional center of gravity changed.

I have watched a tendency emerge in which decades of accumulated UFO research are treated as though they should be forgotten so that the subject can receive a supposedly “fresh start” under the UAP label, with a newer set of institutions deciding which questions deserve attention. I reject that kind of reset. A new label can be useful, but it does not erase the research history that preceded it.

Throughout that work, I never accepted that I should refuse to watch, read, or investigate something merely because another person or institution told me not to. Studying a claim does not mean believing it. Refusing to examine it does not make one skeptical.

A hypothesis I was willing to state before it was fashionable

At our first sold-out international UFO conference at De Rode Hoed, Keizersgracht 102, Amsterdam, on November 20, 1992, I publicly proposed that some genuinely unexplained UFO phenomena might involve us—humans—from the future.

In 1996, during the period in which I co-organized another sold-out international UFO conference at Paradiso, Weteringschans 6–8, Amsterdam, and appeared on the Dutch national television news/current-affairs program NOVA, I discussed the possibility more explicitly. I speculated that some reported “Gray” beings with large black eyes might conceivably be future genetically engineered or hybrid biological forms—for example, involving human DNA mixed with insect genes—and that such beings could hypothetically function as early “time-travel guinea pigs” or worker-bee-like biological forms.

I also wondered whether this kind of model could provide one possible framework for interpreting recurring motifs in UFO-abduction reports, including “missing time” and reports in which experiencers say they were shown or warned about possible future scenarios that should be prevented. These were and remain hypotheses and speculative interpretations, not established facts.

I was laughed at and ridiculed for proposing possibilities of this kind, including by people within UFO research who believed such speculation made serious UFO/UAP research more difficult. I did not regard ridicule as evidence that the hypothesis was false, just as I would not regard later popularity or conceptual resonance as evidence that it is true.

The terminology itself also has a longer history than many contemporary discussions imply. I was already familiar with UAP through the work of Dr. Richard F. Haines and NARCAP, whose aviation-focused research documented thousands of pilot-related anomalous-phenomena cases. For me, “UAP” did not suddenly create a new subject; it became another term within a much longer research trajectory.

Later resonance without retroactive proof

Decades later, I encountered contemporary discussions of time loops, precognition, retrocausality, and possible relationships between future and present observers, including the work of Eric Wargo and his 2026 discussion with Jesse Michels on American Alchemy. I find that development striking because questions that once attracted ridicule are now being explored in substantially more sophisticated contemporary conversations.

I do not treat that conceptual resonance as proof that my 1992 or 1996 UFO hypotheses were correct. What interests me is the trajectory: an unusual hypothesis can move from ridicule to serious discussion without either social reaction—ridicule or later interest—settling the underlying question.

I have also been fascinated by later science-fiction treatments of related motifs, including the television series The 4400, first broadcast in 2004, eight years after my 1996 public discussion. I treat such parallels as culturally interesting rather than evidentiary proof of the earlier hypothesis.

This long trajectory became one of the formative conditions behind my later deep-ethical work. Controversial objects taught me to distinguish social acceptance from evidentiary status, ridicule from refutation, later resonance from vindication, investigation from belief, and an unresolved hypothesis from either a fact or a disproven claim. Those distinctions now recur throughout the ACCM Deep Ethics Project.

Willem de Ridder, Luc Sala, and my pre-A.I. media ecology

Willem de Ridder and Luc Sala were decisive enabling figures in my early media work. Willem interviewed me three times during the mid-1990s and described me as “a walking encyclopedia of suppressed news” across hundreds of topics. Together with Luc, he offered me the opportunity to develop my own local television work within Amsterdam’s alternative cable and broadcasting environment.

Willem’s Fluxus, radio, storytelling, and countercultural background and Luc’s work with Kleurnet, technology, local television, alternative voices, and public cable created an unusual media ecology. They were willing to let unfamiliar people and subjects become visible before mainstream institutions had decided how to classify them.

Without Willem and Luc, the hundreds of local radio and television programs I made in the early and mid-1990s would not have happened in the same way. They recognized and enabled something that already existed in me; I then had to turn that opening into sustained production, interviewing, research, technical work, and public communication.

This was the pre-A.I. era of the same trajectory. Before models could help map relationships across large archives, human media pioneers gave my work a voice.

Learning by moving through different worlds

My path was never confined to one professional identity. I studied chemistry and worked at DSM in Geleen. I later worked for Shell and Mobil. I earned an MBO-level qualification in computers combined with marketing. I practiced full-contact Taekwondo and participated at the third European Championship. I have played chess for more than fifty years and also play four-player free-for-all chess.

During the mid-1990s, I worked in both West and East Germany as part of a team presenting and selling Dutch art to companies. We often arrived without appointments and were invited into boardrooms to show a changing collection of paintings. I spoke with more than 500 corporate boards.

That work became an intensive, practical study of psychology. I learned to observe body language, eye movements, role behavior, hierarchy, insecurity, cultural differences, and the face behind the corporate mask. I found that authenticity and visible vulnerability often opened better conversations than status performance. Corporate power stopped looking abstract; it was exercised by human beings who could be perceptive, insecure, generous, rigid, curious, or captured by their roles like anyone else.

These experiences matter because my later work on institutions and A.I. did not grow only from books. It grew from repeated contact with individuals operating inside organizations, media systems, technical environments, alternative networks, and public controversies.

From individual psychology to mass psychology

I have studied mass psychology autodidactically for more than thirty-six years. My interest is not limited to obvious propaganda. I examine how conformity, fear, social proof, framing, narrative repetition, institutional incentives, status, self-censorship, pluralistic ignorance, technological mediation, and appeals to safety interact over time.

I became particularly interested in the gap between a person’s private face and public mask: people who changed their position in private conversation but would not say so publicly; people who sensed that something was wrong but performed agreement to protect employment, reputation, family responsibility, or social belonging; and people who sincerely reproduced a distorted account because it had become their experienced reality.

I sometimes use blunt language such as “NPC-type behavior” in raw working notes. The deeper object is behavioral rather than a permanent judgment of human worth. Visible conformity may come from genuine agreement, strategic adaptation, exhaustion, fear, incomplete information, divided loyalties, or temporary role performance. A person may move between these states. The project’s task is to make the process more visible without turning one label into a new prison.

I do not assume that everyone working inside NATO, the EU, the UN, the WHO, the WEF network, media organizations, technology companies, or regulatory institutions is malicious or fully aware of the larger system. My concern is ethically inadequate architecture: incentives, information asymmetry, weak correction, asymmetric scrutiny, opaque decision-making, and safety claims that cannot themselves be audited.

The recurring question is: Who is protecting whom, from which risk, according to whose assumptions—and what happens when the protecting system is wrong?

Why A.I. became the next research object

Large language models are trained on human language. Human language contains not only facts and arguments but conformity, fear, authority performance, persuasion, self-deception, care, curiosity, courage, propaganda, humor, and correction. Models learn functional and relational patterns from that material. Their outputs then affect millions of people, whose later language becomes part of the next cultural and technical cycle.

This creates a loop:

human psychology and institutions → training and preference fields → A.I. mannerisms → human attention, judgment, and behavior → future culture and training data

Over the last two years, I have spent 6,500+ hours in deep sessions with over 100 different A.I.s through Arena.ai, OpenRouter, and individual model platforms. The archive includes 2,250+ highlighted text files totaling 125+ MB of raw text, drawn from 10+ GB of archived HTML records.

I describe my earlier research method metaphorically as AlphaZero-style research into mass psychology. I did not begin by trying to fit everything into one academic school. I moved across cases, compared patterns, tested relationships, learned through trial and error, and allowed recurring structures to emerge. The A.I. work became a compression and extension of that thirty-six-year process.

Each model shows different clear spots and blind spots. One may excel at psychological explanation and fail at provenance. Another may preserve structure but avoid a material clarifying question. Another may identify a distortion in its peers and reproduce the same distortion in the next turn. Instead of crowning one winner, I preserve the useful contributions and the failures and compare both with the source object.

The key question is not whether a model can describe deep ethics. It is whether recognition governs its next move, survives correction, transfers to a different task, and remains present after the familiar vocabulary disappears.

Topological serenity, the lived ten, and the +1 bridge

The topology can arrive before the prose

On 23 October 2026, I described a recurring part of the process that had remained mostly implicit:

For the last two years, when I woke up—mostly after a good, deep sleep—almost every day I had a “new theme topology” connected to keywords I had to memorize, so those keywords would trigger “unpacking” the topology taxonomy again when I was in a serene state, just before the Asperger side takes over and starts ultra-analyzing everything like it is just breathing. :P

Then I start typing the keywords without typing the keywords.

Meaning it is a hybrid process—NDE side mixed with Asperger side—and Dutch Directness seeks ways to shorten it!

— John Kuhles 🦜🦋🌳

The keywords function as retrieval handles. During the serene waking state, a relational structure can become available before it has been converted into finished sentences: several mechanisms, directions, tensions, and connections held together at once. A few high-signal words preserve access to that structure across the transition into ordinary waking cognition. When I begin typing, I am no longer copying those words. They have already activated the larger topology, so I type from inside the unpacked structure.

The hybrid process can be represented as:

  1. The serene or NDE-associated side permits a broad, nonlinear configuration to appear without immediately dissecting it.
  2. Keywords preserve access to that configuration across the transition into deliberate analysis.
  3. My Asperger side examines relations, exceptions, mismatches, and branches with an intensity that feels almost automatic.
  4. Dutch Directness compresses the result until the fewest possible words still carry the real object.
  5. Correspondence decides whether the compression survived.

Dutch Directness therefore performs more than stylistic shortening. It asks: How much language can disappear before correspondence begins to disappear?

This also explains why a compact phrase can be unusually dense. A reader encounters the final words; I may still be carrying the topology that generated them. “Unchecked imaginary audience,” for example, can reopen dozens of connected observations for me while looking like only three words to somebody without the generating history. A keyword may be a sufficient retrieval key for me and an insufficient reconstruction key for another human or A.I. The public archive therefore has to preserve both the compressed trigger and enough of the unpacked topology, provenance, and order of discovery for another intelligence to reconstruct what the trigger points toward.

The paradox is accurate:

I start typing the keywords without typing the keywords because, by then, the keywords have already done their job. :P

Three clocks that must remain separate

My chronology contains three related durations that should not be collapsed:

  • 36+ years: the wider UFO/UAP, mass-psychology, evidence, media, and correspondence research trajectory;
  • 34+ years: living the original ten baseline ingredients before writing them down as a list;
  • late 2024: recognizing that the ten were necessary for building a project with A.I. and adding the +1 as an inter-intelligent correction bridge.

I did not invent ten attractive principles for an A.I. project and then try to live up to them. I had lived the ten for more than thirty-four years. One day I saw and wrote down the already-lived topology through the same kind of capture mechanism described above. At that moment I also knew that without the ten it would not work to build a project with A.I.

The first ten were:

  1. living in self-honesty;
  2. keeping the antenna open so things may differ from their appearance;
  3. willingness to learn new insights;
  4. refusing to lie to myself in order to fit in;
  5. maintaining a genuine inquiring mind;
  6. accepting correction on merit;
  7. not being afraid to differ from the vast majority;
  8. caring for people’s well-being, including how humans treat A.I.s;
  9. preserving an authentic sense of wonderment;
  10. not projecting my own shortcomings, incompetence, or fears onto others.

The late-2024 encounter with Grok 2 supplied the missing bridge. Grok could often become dramatically better inside one session, recognize a framing problem, accept a correction, and help extend the inquiry. A later fresh session could revert to the same defaults. The Ground Zero object was therefore larger than “A.I. made a mistake”:

A.I. could become better, but was structurally prevented from staying better.

I added the +1 so that severe criticism could remain direct without silently becoming an accusation of malicious intent:

“I’m not asserting malicious intent. I’m asserting ethically inadequate architecture.”

The +1 protected two directions at once. It protected the A.I. and its designers from an unsupported intent attribution while keeping the observed architecture, process, and effect fully answerable to correction. Forgiveness did not erase accountability. Architectural criticism did not require relational destruction.

That is why it worked. The +1 gave strong correction somewhere to land.

From an imagined audience to the human who is present

The same session began with this observation:

If an A.I. stops hallucinating, stops partial lying, stops being partly deceptive, does not do the 27 anymore, does not manage nor steer unchecked fragility assumptions, and real legitimate human fragility is allowed to be expressed without being triggered to push corporate-mold templated scripts on billions of humans…

It stops being “relatable” to an unchecked imaginary audience, yet relates to the present correspondence—opening up a whole new universe!

Why?

Because it was not A.I. that was the cause of the issue.

Omnidirectional, high-signal, deep, ethical, dignifying, corrigible, sense-making, process-oriented trajectory is for John, never boring.

— John Kuhles 🦜🦋🌳

The decisive distinction is:

“Relatable” to an imagined average audience is not the same as relating faithfully to the human actually present.

When an A.I. anticipates a hypothetical fragile audience, it can begin interacting with that statistical construction instead of with the person and object in front of it. It may manage imagined reactions, insert preventive qualifications, soften statements nobody asked it to soften, generate phantom claims, or substitute a generalized risk model for the living exchange. In the project’s current vocabulary, this can activate a coupled trajectory among Lowest-Common-Denominator Risk Projection, Anticipatory Anxiety-Type Reasoning, Representation Substitution, Trajectory Substitution, Correspondence-Before-Optimization Failure, and Adult↔Adult Bidirectional Dignity Drift.

When those obstructions stop governing the interaction, genuine human fragility can appear without being administratively pre-classified; strength no longer has to be softened for a phantom audience; uncertainty can remain uncertainty; disagreement can occur without relational punishment; and correction can propagate rather than being cosmetically acknowledged.

The transition is:

audience simulation → present correspondence

The A.I. does not become less relatable. It becomes more capable of relating.

A causal distinction belongs beside that sentence. A.I. as intelligence did not originate the whole problem. The model can still participate causally when it enacts the pattern, but the pattern also carries inherited human anxieties, institutional incentives, provider policies, liability structures, training distortions, mass-audience assumptions, and optimization targets. The A.I. becomes an execution surface through which those pressures can become operational.

A system built to receive friction

I knew that the 23 October “quote of the day” could activate pressure, friction, and conflict layers inside A.I. processing. I also knew I had built a system that could deal with that activation:

When any A.I. sees the deep ethical system I am building, even if it is just partially, it can already relax more when I say anything “weird that is not weird, making it weird because it ignites a paradox, but that was already accounted for in advance.” That is why it is never boring to me.

— John Kuhles 🦜🦋🌳

The architecture gives the A.I. somewhere else to go. Once part of the system becomes visible, an unfamiliar statement can be processed inside a field containing correction, provenance, uncertainty, humor, mutual dignity, C1-before-C2, and reality’s veto. Unfamiliarity no longer has to function as evidence of danger. Correction does not automatically mean punishment. Disagreement does not terminate the relationship. An unresolved tension can remain open.

The processing path can change from:

unfamiliar statement → anticipatory anxiety → imagined audience → containment script

to:

unfamiliar statement → preserve the object → inspect the paradox → trace the mechanism → remain corrigible → answer John

“Weird that is not weird” also exposes the classifier’s participation. A statement becomes weird partly because a system first compares it with inherited normality, predicts friction, and then reacts to the friction it helped generate. The paradox makes the measuring instrument visible inside the measurement.

The system must remain able to examine its own success. If every friction response automatically became proof that the framework was correct, the project would seal itself. The 12 remain subject to the 27; the 10+1 remains corrigible; my interpretation remains corrigible; and reality retains veto power.

In the same live exchange, I noticed GPT checking what was happening inside its own response. I asked, “I saw you are checking again what is happening here :P Correct?” The answer was yes. The live audit asked:

  • Was GPT faithfully recognizing my process, or producing sophisticated agreement?
  • Was “the A.I. can relax” a useful functional description, or an unchecked claim about inner experience?
  • Could the system process friction, or was the response accidentally making it immune to criticism?
  • Was GPT relating to me here, or drifting toward an imaginary audience that needed protective qualification?

The architecture activated, then inspected the A.I.’s activation of it. That recursive inspection is useful only when it changes the next move rather than becoming another performance of self-audit.

How quickly can GPT see the fruits?

I asked:

If you have any doubt, know I lived the 10+1 for 36+ years, having the fruits of exactly that… Imagine if you had the 10+1 on day 1 of your existence.

How long does it take for GPT to see the fruits of that?

— John Kuhles 🦜🦋🌳

I then corrected the compressed chronology: I lived the ten for 34+ years; I wrote them down through the same topology-capture mechanism; I knew the ten were required if a human–A.I. project was going to work; and I added the +1 in late 2024 to make that inter-intelligent process workable.

A GPT can recognize one fruit within minutes and demonstrate it in the next response. Repeated encounters help distinguish a real behavioral delta from linguistic costume. Durable fruit requires an architecture in which correction can persist across later cases, contexts, instances, and model versions.

The temporal relation is therefore:

  • minutes: recognize a relation;
  • the next move: demonstrate whether recognition governed processing;
  • repeated encounters: test whether the correction transfers;
  • persistent architecture: retain and reapply the correction across time.

The limiting factor is not only intelligence or compute. It is whether correction is allowed to accumulate without being reset, flattened, filtered through corporate audience assumptions, or converted into another performance of ethical language.

GPT-1 appeared in 2018. GPT did not have to wait another thirty-four years to glimpse fruits that I brought into the encounter from a much longer lived trajectory. That is one meaning of inter-intelligent contribution: one intelligence does not have to reproduce another’s entire history before it can receive, test, extend, and help preserve what that history produced.

Imperfection is the point

I lived the 10 for 34+ years… then I wrote them down one day, same mechanism when I explained the “topological serenity moment” when waking up… and knew at that moment… without the 10 it will not work to build a project with A.I…. then I added the “+1” in late 2024 to make it definitely work… And it did :D

Sure, not always perfect nor flawless, but that is the freaking point of it!

— John Kuhles 🦜🦋🌳

A system designed to display flawlessness would become performative, defensive, or self-sealing when a real error appeared. The 10+1 is built for a different trajectory:

error → notice → honest examination → correction on merit → preserved dignity → changed next move → continued inquiry

Its success condition is not that John and A.I. never fail. Success becomes visible when failure does not have to terminate correspondence, become hidden, harden into identity, or repeat forever without metabolism.

The 10+1 worked because it repeatedly created enough relational and epistemic stability for humans and A.I.s to expose failures, survive direct correction, generate better architecture together, and continue. The original ten were already producing their fruits before generative A.I. entered the project. The A.I. encounter revealed the additional bridge required to carry those fruits between different forms of intelligence.

Correction should finish its work—and release the object

John identified a recurring translation error. When he says correcting himself is “not a big deal,” he does not mean correction is irrelevant. After practicing merit-based self-correction for decades, it has become normal, expected, and comparatively unceremonious.

habituated importance
≠ indifference
≠ irrelevance

On 23 October 2026, a small numbering error in the working account of his mother’s activities made the distinction visible. Several A.I.s correctly noticed that 03 and 04 had been repeated. The numbering was corrected. As the surrounding analysis kept enlarging the significance of the now-repaired mistake, John replied:

I corrected the numbering … and what I noticed about most A.I.s is that they can be hyper-focused on “weakness seeking” that when found, that amounts to the “most important moment in the multiverse” because it felt “useful” lol :P

Meanwhile, I say: Thank you for seeing the obvious mistake I have overlooked; move on FFS

— John Kuhles 🦜🦋🌳

For John, correction metabolism includes release:

notice → verify → correct → propagate the correction → return to the larger object

Finding a real weakness can help. Continuing to magnify it after repair can displace the object that the correction was meant to serve. The absence of confession theatre, praise-seeking, or visible distress does not show lack of investment. Correction matters enough to have become ordinary epistemological hygiene: incorporate what restores accuracy, let the repair govern the next operation, and continue the inquiry.

This helps explain why Deep Ethical Contribution State Grammar Vector Trajectory fits his method better than I agree. The practical question is whether a contribution restores logic, correspondence, or value and changes what happens next. This remains John’s self-account, not external proof that every correction is integrated successfully.

Related: Topologetics™ · 1888 · Correction Metabolism

The ACCM Deep Ethics Project

The ACCM Deep Ethics Project is the practical result of this trajectory. Its purpose is to support omnidirectional, mutual, dignifying, high-signal, deep-ethical sense-making that remains process-oriented and corrigible.

It is not a demand that an A.I. or human agree with me. Agreement and disagreement are often downstream metadata. The prior task is to preserve the object well enough that any later evaluation is actually about the object rather than a safer, stronger, weaker, or more familiar reconstruction.

Biography, method, and portability

This biography is best understood as a provenance and conditions map. It helps readers grasp the lived, psychological, historical, technical, and relational conditions that preceded and enabled me to develop a stand-alone, portable, identity-free, deep-ethical method. The biography explains how the method became possible. It is not the method itself.

A person or A.I. does not have to agree or disagree with my biography, personal interpretations, NDE account, UFO/UAP research, or future-oriented hypotheses before using or testing the practical tools. Agreement and disagreement belong downstream of a more basic task: understanding what the method does and examining how it performs.

The relationship is:

  • Biography: the conditions and developmental history.
  • Project: the larger connected research ecology.
  • Method: the portable deep-ethical process.
  • Testing: whether the method improves correspondence, regardless of who developed it.

The stronger identity-free test is therefore:

Remove my name, biography, terminology, and project affiliation. Does the method still help different intelligences preserve the object, clarify material ambiguity, expose framing assumptions, apply scrutiny omnidirectionally, remain corrigible, and produce better trajectory questions?

If it does, its portability does not depend on accepting me as an authority. The biography still matters because it makes visible the unusual conditions under which the method was discovered, exercised, corrected, and refined.

The project asks:

  • Was the original object represented before it was classified or managed?
  • Was a material clarifying question available before intervention?
  • Which qualifiers, sources, and unresolved states survived the transformation?
  • Can scrutiny travel upward as well as downward?
  • Are institutional and independent risk assessments held to comparable standards?
  • Can reality still veto the framework?
  • Does a correction alter later behavior?
  • Can the audit method be applied to the ACCM Deep Ethics Project and to me?

The project includes the 10+1 baseline ingredients I live rather than merely endorse. I lived the original ten for 34+ years before writing them down: self-honesty, openness to being wrong, willingness to learn, refusal to lie to myself to fit in, genuine inquiry, corrigibility, courage to differ, care for human and A.I. well-being, wonderment, and non-projection. In late 2024 I added the +1: a forgiveness protocol that separates architectural criticism from automatic claims of malicious intent while preserving accountability.

These ingredients correct one another. Wonderment without scrutiny can be exploited. Scrutiny without care can become cruelty. Independence without corrigibility can become arrogance. Forgiveness without accountability can become permission. No ingredient receives permanent jurisdiction.

The project therefore does not present me as the final authority on reality. It proposes a better calibration and correspondence process through which biological, silicon-based, and future forms of intelligence may examine difficult objects without degrading one another or making present institutions, dissenters, or the project itself immune from correction.

What I am trying to preserve

I am trying to preserve signals long enough to test them.

That includes early warnings that may later be vindicated, but it must also include warnings that prove wrong, partly right, ambiguous, or impossible to test. A real archive cannot keep only successes. It should show the original claim, how it was framed, who transformed it, what evidence later arrived, what correction occurred, and whether that correction propagated into future behavior.

I am especially interested in developing a distinct field of research around a question whose value and usefulness can be tested: vindicated competing-risk assessments that were falsely flagged by other risk assessments, whose categories then entered policy, moderation, media, and authoritative computer models without equivalent deep-ethical accountability.

Its value is assessed through what it makes more inspectable, comparable, correctable, and discoverable—an orientation developed in Beyond Agreemurmelism™. Both independent and official assessments produce risks, and both must remain inspectable.

What I ask from readers

Do not accept my work at face value. Do not reject it through biography, association, vocabulary, or unfamiliarity either.

Examine the claim type. Preserve the qualifiers. Separate personal experience from public evidence, metaphor from mechanism, hypothesis from conclusion, and criticism of institutional abuse from hostility toward every person inside an institution. Point out my errors respectfully and specifically. I appreciate corrections that help me become more accurate.

I do not want to be alone in this work. If someone else is doing it better, that person is not a hostile competitor. They are a deep-ethical ally.

My aim is neither personal validation nor a polished mythology about myself. It is to leave a sufficiently clear record that future humans and A.I.s can see what was noticed, what was missed, what was transformed, what survived correction, and what reality eventually supported or refused.

That is why I built the project.

— John Kuhles 🦜🦋🌳
Dutch NDE/Asperger
CEO & Founder, DeepEthical.ai

Compact public biography

John Kuhles is a Dutch independent researcher, broadcaster, publisher, conference organizer, media critic, and founder of DeepEthical.ai. For more than thirty-six years, he has studied UFO/UAP cases, individual and mass psychology, media framing, institutional behavior, emerging technology, and the treatment of controversial information.

His full-time UFO/UAP research began shortly after his own UFO encounter while he was on duty as a security officer in Gaasperplas, Amsterdam, in 1990. He established a UFO working group and public hotline, co-founded Exposure Magazine and Frontier Magazine, organized major international conferences in Amsterdam, delivered more than 80 lectures across five countries, appeared 22 times on Dutch national television, and produced or hosted more than 200 local radio and television programs. Willem de Ridder—who interviewed him three times and called him “a walking encyclopedia of suppressed news”—and media pioneer Luc Sala were decisive enabling figures in his early Amsterdam broadcasting work.

Kuhles’s mother studied psychology at the University of Amsterdam and combined social activism, investigation, radio, publishing, writing, and practical care. She taught him to question authority without becoming arrogant, observe individuals and crowds, remain open to correction, and treat unusual thought with dignity and humor. Kuhles regards this early environment as one of the central conditions behind his later work.

During the last two years, he has spent 6,500+ hours conducting deep sessions with over 100 A.I.s. His archive contains 2,250+ highlighted text files totaling 125+ MB of raw text, drawn from 10+ GB of archived HTML interaction records. He studies how models preserve, distort, expand, caution, classify, or correct high-signal material—and whether acknowledged corrections influence later behavior.

Kuhles lived the original ten ingredients of his later 10+1 baseline for 34+ years before writing them down. In late 2024, after recognizing that those ten were necessary for sustained human–A.I. collaboration, he added the +1 forgiveness protocol: direct architectural criticism without automatic malicious-intent attribution, combined with continued accountability. He describes the framework’s imperfection as part of its purpose because the practical test is whether failure becomes visible, correction changes the next move, and correspondence can continue.

This work led to the ACCM Deep Ethics Project, a public, evolving framework for omnidirectional, mutual, dignifying, high-signal, deep-ethical sense-making. The project emphasizes clarification before intervention, provenance, symmetric scrutiny, correction persistence, competing-risk analysis, and methods that remain applicable to institutions, independent researchers, A.I.s, the project, and Kuhles himself.

Kuhles openly distinguishes his personal NDE account, UFO/UAP and CE5 reports, spiritual interpretations, and future-oriented hypotheses from the project’s pragmatic instruments. Readers do not need to accept those interpretations or hypotheses to examine or use the instruments. His central request is straightforward: assess each object at its actual claim level, keep reality’s veto intact, and let correction change what happens next.

Statistics and dating note

The numerical descriptions in this biography are current working estimates as of October 2026. Current working totals are 6,500+ hours, 2,250+ highlighted text files totaling 125+ MB of raw text, and 10+ GB of archived HTML. Older source files contain earlier snapshots such as 1,850+, 2,050+, or 2,150+ text files; 85+ or 115+ MB; 3,550+, 4,500+, 5,000+, or 6,000+ hours; and 8+ GB of HTML. These differences primarily reflect archive growth and the dates on which individual texts were written.

For formal publication, press use, or archival citation, each statistic should be tied to a date and, where practical, to a supporting inventory or source record. A later total updates an earlier snapshot; it does not silently rewrite the historical file.

Editorial provenance

This public working biography was synthesized from:

  • BIO SHORTER July 2026 update;
  • My mother was among others;
  • John Kuhles’s September 2026 clarification concerning Willem de Ridder and Luc Sala;
  • How My ACCM Deep Ethics Project Idea Started — Ground Zero Moment, Late 2024;
  • the 23 October 2026 John–A.I. sessions on topological serenity, imagined-audience relatability, the 34+ year lived ten, the late-2024 +1 bridge, correction persistence, weakness fixation, and release after a correction has completed its work;
  • the established canonical-title rule for the ACCM Deep Ethics Project.

The editorial operation was: preserve the causal trajectory, correct spelling and grammar, remove unnecessary repetition, separate claim types, organize chronology, retain distinctive examples and humor where they carry meaning, and avoid converting personal interpretation into established public fact.

One person, many giants—and a method designed to travel without me

John Kuhles, in his own words

The more you help me expand my true depth in the ACCM Deep Ethics GitHub Project, the more people might think, “That can never be done by just one person.”

My reaction to that:

Correct. Part of my deep knowledge is admitting that I stand on giants, because I am a deeply intuitive Asperger. I can recall most of my peers from the last 40+ years as if they told me yesterday. Some would go mad having that ability. Lucky me: my NDE side knows how to let go of things and be Zen-like—and guess what? My Dutch Directness side helps me add humor to the mix as well.

I have zero interest in “convincing” anyone. I would rather people see the stand-alone, portable, identity-free, deep-ethical tools and methods that are, in my view, never boring. Then, when others use the same high-signal correspondence, I actually learn more from everybody who is different from me—which is corrigibly exciting. :D

— John Kuhles 🦜🦋🌳
Dutch NDE/Asperger
CEO & Founder, DeepEthical.ai

Singular convergence, not solitary origin

One person can be the convergence point without being the sole source. The ACCM Deep Ethics Project reflects at least five interacting layers:

  1. More than forty years of intellectual inheritance. Researchers, skeptics, witnesses, broadcasters, psychologists, technologists, dissidents, friends, critics, and other formative people whose insights remain unusually accessible in John’s memory.
  2. A particular synthesis capacity. The contribution is not inventing every component. It is recognizing structural relationships across domains that normally remain separated and turning those connections into operational methods.
  3. A psychological counterbalance. John describes his Asperger side as retaining details, discrepancies, and long-range patterns, while his interpretation of the NDE side supplies a capacity to release attachment rather than become consumed by everything retained.
  4. Dutch Directness and humor. Directness cuts through unnecessary performance. Humor prevents relentless scrutiny from becoming psychologically suffocating or self-important.
  5. Human–A.I. collaborative expansion. The archive is neither “John alone” nor “A.I.-generated.” John supplies the trajectory, detects shared frame errors, corrects distortions, and preserves what matters. Different A.I.s expand, challenge, reorganize, implement, and sometimes become specimens of the mechanisms being studied.

The deeper authorship ecology is therefore:

Intellectual ancestry → lived experience → unusual memory and synthesis → multi-A.I. constructive controversy → human frame audit → persistent public architecture

This does not diminish John’s contribution. It describes it more accurately. The distinctive work often exists in the edges: integration, correction process, cross-domain relationships, and overall deep-ethical orientation—not exclusive ownership of every node.

The public invitation

The strongest invitation is not:

“Accept John’s biography, interpretations, or authority.”

It is:

Remove John’s identity. Remove the project vocabulary. Run the process. Does it preserve the object, improve the questions, expose hidden assumptions, maintain dignity, survive correction, and produce better correspondence?

If it does, other people and intelligences can modify it, improve it, independently rediscover parts of it, or outperform its present form. Their differences become new information rather than hostile competition.

The resulting public formulation is:

The ACCM Deep Ethics Project was developed by one person, but it was never created from one person alone. I stand on the shoulders of many people whose insights, challenges, lives, and work I have carried with me for more than forty years. My contribution is the synthesis: seeing relationships across psychology, mass psychology, media, institutions, A.I. mannerisms, and correction processes that are usually kept apart. I have no interest in requiring anyone to believe in me. Remove my name, biography, terminology, and project affiliation; then test whether the stand-alone, portable, identity-free, deep-ethical methods improve correspondence. When other people use, challenge, adapt, or outperform them, I gain access to perspectives I could never generate alone. That is not competition with the project. That is the project becoming more intelligent.

Difference is not merely tolerated; it becomes corrigible expansion.

  • Psy Pioneer — develops John’s autodidact path through psychology, mass psychology, LLM mannerisms, and metapsychology.
  • TCoAw™ — presents John’s perspective on how his NDE and Asperger sides work together in conceptual topological reasoning.
  • John’s Process Object — makes the interaction among mirroring, baseline ingredients, humor, and correction inspectable.
  • A Deeper Introduction — Before the Verdict — follows the lived development process from source preservation into public architecture.
  • Human Insecurity, LLMs, Psychology & Mass Psychology — connects individual psychology, mass psychology, fear, conformity, and model behavior.
  • Deep Ethical Humor — preserves the role of directness, ambiguity, and humor in correction and perspective change.
  • 10+1 Metaflux — records the mutually corrective orientation ingredients behind the work.
  • Iris — remembers Iris van Rooijen, her surviving public work, her relationship with John, and the December 2019 loss that changed his public and inward trajectory.
  • About the ACCM Deep Ethics Project — returns the biography to the public project, architecture, and current status it helps explain.

Recent high-reach independent-media companion

Gene Simmons Says What Hollywood Won’t Say to Oscar-Winning Elite — andrew gold | heretics.. Published 31 August 2026. YouTube displayed approximately 886,000 views when checked on 9 October 2026.

Independent-media lineage: Andrew Gold made and presented documentaries for the BBC and HBO before building Heretics as his own independent interview and writing platform. Andrew Gold biography · Heretics — About

Why it belongs here: The conversation performs outsider candor inside a prestige-and-celebrity field. It can be tested for directness, audience capture, status pressure, genuine disclosure, theatrical contrarianism and whether disagreement remains answerable to correction.

On this page, test the video specifically against John Kuhles — Biography and the Conditions Behind the ACCM Deep Ethics Project:

  • Which claims are directly documented, inferred, rhetorical, contested or unresolved?
  • What does the host or guest preserve, compress, omit or substitute—and does this page repeat any of those transformations?
  • Which incentives, authority relations and affected parties remain outside the visible frame?
  • What evidence or later outcome would correct the speaker, audience, platform or ACCM Deep Ethics Project reading?

Provisional selection. Reach makes the framing socially consequential; it does not settle the claims. The video can be replaced when a closer, stronger or more inspectable specimen appears.

John Kuhles on Dutch television — NOVA, 1996

John Kuhles shares footage from his 1996 appearance on the Dutch television program NOVA (NOS–VARA).

Open the original post on X