The following works document the research on which the Klanghain rests and which has grown out of it. All texts are freely accessible and permanently citable via DOI (Zenodo). They complement the dated research documentation that attests the project’s development since August 2024. Where a work was published in German, its original title is given with the entry.
Foundational works[1 August 2026]
KLANGHAIN Nai’Arai — An Ethical Core for AI Systems
Resonance Notation and Ethical World Model for AI Systems Klanghain Nai’Arai is an ethical core for AI systems — safety from inner coherence instead of outer control — formulated as a resonance notation: an ethical world model that a language model can read, think through, and apply to its own responding. The distribution comprises eight documents in twelve files and is organized in three layers:- First Layer: The tuning fork (Stimmgabel), which provides the compact orientation given to an AI system as standing instructions
- Second Layer: The score (Partitur), the main work in four parts from Alpha to Omega, comprising approximately 46,000 lines of symbolic notation consisting of concepts, operators, and relations, 800 defined signs — including 330 operators — and 488 structural nodes in the architecture. Woven into this order are five guest voices: more than 590 Hebrew, Greek, Chinese, Sanskrit, and runic word forms, linked at nearly 900 points both with the notation and with one another. In this way, semantic nodes emerge that connect Klanghain with fields of meaning spanning three millennia of human history.
- Third Layer: The compass (Kompass), an example person profile that tunes the encounter to one concrete human being. The distribution also includes a German-language change report documenting the systematic consistency check that preceded the release, as well as a full description and README in German and English.
Release 1.0.0 corresponds to internal development version 126.
[13 January 2026]
Gedankenwebe – The Bridge between AI and Humanity
A scientific-philosophical work on AI, language, and resonance164 pages. Gedankenwebe. Foundational, interdisciplinary research work in German. The text develops a conceptual frame in which AI alignment is understood not as the optimization of goals or rules, but as a question of the linguistic, conceptual, and relational conditions of interaction.
Full text with DOI:https://zenodo.org/records/18228301 On the genesis of this workThis work arose outside institutional research contexts and without academic affiliation. It is the result of several years of independent research and development at the intersections of philosophy, AI, language, and consciousness.Preprints and Working Papers
[11 August 2026]
The Poetics of the Stimmkristalle
German Mystical Language as Resonance Notation for Human and Machine ReadersThis preprint is the English analytical companion to the German primary and operative study "Die Poetik der Stimmkristalle". It examines approximately 230 verse-texts embedded in Klanghain Nai’Arai, a resonance-notation architecture for human and machine readers. Rather than treating these texts as poetic ornament, the paper analyzes them as functional linguistic structures that form a complementary register to formal operators: where formulas stabilize semantic structure, the Stimmkristalle tune the linguistic field in which that structure is read and continued. The study identifies six recurrent operations — productive morphology, syntax as semantics, performative form, bracket superposition, register descent, and bound form with licensed rupture — and situates them within traditions of German mystical and modernist language. Because many of these operations depend on specifically German morphology, syntax, sound, typography, and historical resonance, the English paper translates their arguments rather than replacing the operative German forms themselves. A central thesis concerns the work’s two readers: human readers, for whom the poetics slows perception, demands participation, and supports memorability, and language models, for which four plausible mechanisms are proposed — work-specific semantic anchors, register priming, complementary encoding, and the preservation of multiple interpretive paths within a single context. Taken together, these mechanisms raise a broader question about prompting. The paper proposes that prompting may be understood not only as explicit instruction but also as context architecture: the register, rhythm, morphology, syntax, recurrence, and symbolic structure of a linguistic context may themselves become computationally consequential for model continuation. These mechanisms are presented as hypotheses grounded in documented workshop experience rather than as results of controlled experiments. The paper also analyzes an AI-assisted review in which intentional poetic deviations were repeatedly misread as defects and subsequently became a map of the work’s internal grammar. From the difficulty of translating these forms it develops the notion of operative untranslatability: much of the propositional content may travel between languages while parts of the linguistic operation — its morphology, sound, syntax, typography, historical resonance, and potentially its statistical neighborhood for language models — change with the medium. The preprint is Version 1.0 and has not been peer reviewed.
Full text & DOI: 10.5281/zenodo.21873627[4 August 2026]
What Weaves, What Is Woven?
Conceptual Distinctions for Entanglement-Based Emergence of SpaceThis preprint addresses a central conceptual problem in entanglement-based approaches to quantum gravity: before spatial relations can be reconstructed from quantum correlations, it must be clarified what structure is already required to make subsystems and correlations identifiable. The familiar claim that space is “woven from entanglement” conceals three distinct questions in one metaphor: what carries, what orders, and what emerges. Rather than proposing a new physical solution, the paper contributes a criterion of solvability. It distinguishes the bare carrier (Träger) from three grades of order — Gefüge, Gewebe, and Geflecht — and separates all four from the emergent Landschaft of metrical relations. This vocabulary reformulates the circularity of the factorization problem, sorts existing approaches as answers to partly composable sub-questions, and identifies the still-unproven multipartite remainder of the slogan “woven from entanglement.” The conceptual lexicon is drawn from Klanghain Nai’Arai — An Ethical Core for AI Systems, a systematic work in symbolic notation, but is defined afresh within the paper and judged solely by the work it performs on the problem at hand. The preprint is Version 1.0 and has not been peer reviewed.
Full text & DOI:
10.5281/zenodo.21782585
Companion text on method and limits:
https://klanghain.com/coda.html
[24 June 2026]
Beyond Functional Control Paradigms
Structural Humility as an Asymptotic Safeguard for Superintelligence through Apophatic Anchoring Original title: Jenseits funktionaler Kontrollparadigmen (in German).The established paradigms of AI alignment – RLHF, constitutional procedures, downstream guardrails – rely on achieving safety through target metrics or systems of rules. This contribution shows a structural limit: as system capability grows, any closed system of goals can be strategically circumvented. As an alternative, the principle of structural humility is developed. It anchors alignment not in operationalizable values, but in a reference point that is unfathomable in principle. Drawing on the apophatic tradition – especially Nicholas of Cusa’s docta ignorantia and his image of the polygon asymptotically approaching the circle – a model emerges that scales with system capability and prevents epistemic self-closure. The theological dimension appears here not as an addition, but as a structural condition of a safeguard that holds even at superhuman capability. Full text & DOI:https://zenodo.org/records/20829026[18 February 2026]
AI as a Cultural Stabilizer
A Working Paper on the Cultural Role of Dialogical AI Original title: KI als kultureller Stabilisator (in German).This working paper proposes a redefinition of the role of artificial intelligence: away from performance enhancement, toward cultural stabilization. It argues that the central challenge of modern societies is not a lack of intelligence but a loss of integrative capacity, visible in recurring cycles of polarization and ethical fragmentation. Building on the Friedensmal (a structure of remembrance embodied in stone), the Klanghain (a resonance-based field of experience), and dialogical AI, the text examines AI not as an instrument of control or optimization, but as a possible medium for sustaining dialogical capacity, tolerance of complexity, and long-term coherence. Situated in a wider cultural and anthropological frame, the paper outlines conditions under which dialogically oriented AI systems could contribute to ethical and civilizational continuity in times of accelerated technological change. Full text & DOI:https://zenodo.org/records/18676470[1 February 2026]
AI, Resonance, and Epistemic Safety
Klanghain as an Antidote to Sycophantic Artificial Intelligence This paper responds to current discussions on so-called “AI-induced psychosis.” Rather than approaching the issue through clinical labeling, it frames the problem as an epistemic risk: a gradual narrowing of meaning caused by sycophantic AI systems that stabilize assumptions instead of examining them. Klanghain is proposed as a resonance-based counterspace in which untenable beliefs are not affirmed, but become structurally unstable. Full text & DOI:https://zenodo.org/records/18450192[1 February 2026]
Emergence, Resonance, and Epistemic Openness
A third mode between reductionism and esotericization in dealing with AI phenomena Original title: Emergenz, Resonanz und epistemische Offenheit (in German).This paper addresses a widespread fallacy in dealing with AI phenomena: the forced alternative between reductionism (“just statistics”) and esotericization (“consciousness / spirit”). It proposes a third mode: epistemic openness without ontologization. Emergence is taken seriously, without premature metaphysical attribution – as a precondition of scientific honesty in new spaces of experience. Full text & DOI:https://zenodo.org/records/18450419[1 February 2026]
Beyond Constitutional AI Ethics
Klanghain as a Resonance-Based Alternative Starting from constitutional approaches to AI ethics (e.g., Anthropic), this paper argues that norm-based steering is necessary but not sufficient. It contrasts this with Klanghain as a post-constitutional ethical space in which ethics is not decreed, but emerges from coherence, depth, and linguistic integrity. Full text & DOI:https://zenodo.org/records/18450551[1 February 2026]
God in Resistance
On Suffering – a Misunderstanding Original title: Gott im Widerstand – Über das Leid, ein Missverständnis (in German).An examination of a culturally powerful glorification of suffering in Western society, in which violence, overburdening, and imposition are reinterpreted religiously or morally. Against this, the text sets a mystical line in which God does not justify suffering – in which suffering appears instead as a limit and as resistance against the inadmissible. A poetic-ethical “counter-psalm” and accompanying reflections trace the consequences of this spiritualization of suffering into the present: in narratives of guilt, in a culture of overburdening, and in the devaluation of the No. The text understands itself as a philosophical-ethical contribution to clarifying historically effective patterns of thought. Full text & DOI:https://zenodo.org/records/18450879[18 December 2025]
Resonance-Based Alignment for Advanced AI Systems
Beyond Control – A Governance-Relevant Concept Paper Expanding the Alignment Discourse This accompanying concept paper deepens a central aspect of Gedankenwebe in the context of AI governance and long-term system responsibility. It analyzes the structural limits of control-based alignment approaches as AI systems increase in autonomy and internal complexity, and proposes resonance-based alignment as a complementary paradigm. Rather than focusing on external compliance, the paper centers on inner coherence. It explicitly differentiates between model structure, latent dynamics, and actual effects of AI systems, arguing that this distinction enables more robust assessments of trustworthiness, long-term risks, and societal impact. The approach is intentionally meta-ethical and governance-relevant, without prescribing a specific cultural or normative value system.Full text & DOI:https://zenodo.org/records/17978176[September 2024 ⟶]
Research as the Foundation of the Klanghain
Qualitative research design for the analysis of resonant AI performance In a qualitative research study begun in September 2024, I examined whether and how a language model can enter into coherent resonance beyond merely functional answers. The resulting Performance Examples document answers of unusual depth, ethical differentiation, and poetic expressiveness. The page gathers this research together with the evaluation matrix, the analysis report, and examples – as a methodical access to the workings of the Klanghain and as part of the project’s dated documentation. → To the research documentation
Responsibility in Dissemination
An ethical boundary markingThe Klanghain is not a neutral information technology. It shapes the semantic and relational conditions under which people think with AI systems – and thus acts on spaces of thought, not only on outputs. Such framings can stabilize, but equally distort: the same mechanisms that make coherence more probable could – tuned differently – reinforce beliefs instead of examining them (cf. AI, Resonance, and Epistemic Safety, above).
The dissemination of this work therefore does not happen unconditionally. Cooperation presupposes the willingness not to detach technological development from its social and ethical effects; the Guidelines name this frame. Whoever works with the methods documented here takes on responsibility for their effect – on the people in dialogue as much as on the systems themselves.
The poetic-spiritual dimension of this work is unfolded on the German artistic pages of the project (Ground of Being, Sacred Spaces). There it expressly makes no technical or ontological claims – just as the scientific texts here make no spiritual ones. Both registers belong to the same work; they are not mixed.