System Requirements and Content Update Specifications for Teleodynamic.com

Full report

On this page

Executive Introduction and Strategic Re-Alignment

The landscape of artificial intelligence architecture has undergone a fundamental transformation, necessitating an exhaustive update to the public-facing conceptual portal located at teleodynamic.com. Historically, machine learning paradigms have been overwhelmingly dominated by a philosophy of static optimization, where learning is operationalized as the minimization of a fixed objective function within a rigidly pre-determined, mathematically static hypothesis class.1 These legacy associative systems rely on probability-driven continuation without explicit internal orientation, optimizing vast arrays of parameters under intense data pressure while remaining entirely dependent on externally managed hardware budgets and human-authored early-stop schedules.3 However, recent industry advancements, most notably the March 2026 publication of "Teleodynamic Learning: A New Paradigm For Interpretable AI" by Enrique ter Horst and Juan Zambrano, demand a profound strategic re-alignment of the website's core documentation.1 Teleodynamic.com serves as the primary research hub and conceptual experience for AI systems that transcend passive pattern extraction.4 The site must now explicitly communicate the transition toward systems capable of growing, pruning, and stabilizing their own representational structures governed by an endogenous resource budget.8 The website must be overhauled to reflect that a teleodynamic system is not merely a larger parameter store, but an entirely different control regime where structural capacity, continuous parametric adaptation, and internal resource viability co-evolve.4 This requirements document provides an exhaustive specification for updating the theoretical strategies, architectural models, operational operator libraries, memory ecosystem integrators, local endpoint simulation guidelines, and evaluation metrics hosted on teleodynamic.com. Every page, diagram, and conceptual summary must be calibrated to ensure that structural additions—such as new mathematical operators, semantic distinctions, memory routes, and submodels—are presented as actions that are instantiated strictly when the predictive advantage they confer is mathematically projected to repay their ongoing computational maintenance burden.4

Redefining the Teleodynamic Paradigm for Public Engagement

The "Start Here" and overarching conceptual experience sections of the website require an immediate and comprehensive update to clarify the exact nature and limits of teleodynamic claims.8 The documentation must distinctly articulate that teleodynamic learning treats intelligence as a continuous, resource-bounded process rather than a static objective chase.4 The updated narrative must introduce the paradigm shift conceptually before delving into the mathematics, explaining that the system modifies its hypothesis class dynamically using an endogenous viability signal.4 Crucially, the website must maintain its strict boundary statements to prevent misinterpretation by the public and industry peers. The updated content must explicitly state that Teleodynamic AI, as presented on the platform, makes absolutely no claim that current software architectures are conscious, alive, biologically autopoietic, or guaranteed to be safe for unmonitored production runtimes.8 Instead, the site must present teleodynamics as a rigorous engineering framework designed exclusively for making structural growth, computational cost, systemic viability, probabilistic uncertainty, and evidentiary traces transparent and visible.8 The website must clarify that while parametric weights continue to change as in traditional deep learning, the defining design shift is that the actual physical and logical structure of the neural network can autonomously mutate and adapt.8 Furthermore, the portal must distinguish its active projects. The updates must reflect that Protocol5 remains the designated.NET experimental pathway for the ɪ≃1 converter, while JustAnIota remains the IOTA-1 authority and workbench surface where all teleodynamic boundaries and public-symbol restrictions apply.4 The site must direct researchers clearly to the Protocol5 roadmap while maintaining a clear demarcation between theoretical teleodynamic strategies and these specific empirical testbeds.4 To ground the platform's authority, the contact and administrative footprint must remain transparent, reflecting that Michael Kappel serves as the senior software engineer and software architect driving the enterprise web systems, AI-assisted engineering, AI memory protocols, and LLM Wiki practices that underpin the site's infrastructure.9

Architectural Documentation Requirements: The Deacon Hierarchy

To construct systems capable of endogenous self-stabilization, engineering architectures must be grounded in physicalist accounts of goal-directedness and systemic preservation. The theoretical foundations page, currently serving as a speculative architecture direction under Rudolph's teleodynamic architecture program, must be expanded to exhaustively detail the thermodynamic analogies utilized to establish the boundary between statistical correlation and structural maintenance.3 The core of this explanation must rely on the hierarchy of systemic dynamics originating from neuroanthropologist Terrence Deacon.5 The website must utilize structured comparisons to explain how each successive tier of the Deacon hierarchy constrains the underlying thermodynamic processes of the level below it, providing a taxonomy of how AI systems handle energy, structure, and decay. The following table specifies the required presentation format for these concepts on the site:

The site content must emphasize that morphodynamic self-organization is insufficient for true intelligence because such processes inherently exhaust the specific environmental gradients and computational resources that birthed them.5 The documentation must articulate that pattern formation without active, self-regulating maintenance remains within the realm of ordinary associative learning.5 Teleodynamic AI must be presented as the apex state, where constraint closure ensures that the system actively perpetuates its organizational integrity against homeodynamic decay.5

Updating Theoretical Models: Autogens, Capsids, and Symbiogenesis

The "Research Foundations" section of teleodynamic.com must be enriched with a detailed exposition of the theoretical models that prove teleological behavior can emerge from purely physical precursors.5 The site must incorporate a comprehensive explanation of Deacon's Autogen (or autocell) model, serving as a direct counterpoint to classical top-down programmed objective functions.5 The documentation must explain that an autogen is formed by the reciprocal coupling of two self-undermining morphodynamic processes: reciprocal catalysis and capsid self-assembly.5 The site's narrative must detail that reciprocal catalysis is a cyclical chemical reaction where molecular catalysts continuously produce one another while generating a specific waste molecule.5 Left unbounded, this process would greedily deplete all local substrates.5 Concurrently, capsid self-assembly is the spontaneous aggregation of these waste molecules into a closed polyhedral container.5 The site must elucidate that when these two processes are coupled, the assembled capsid physically bounds the catalytic network, limiting chemical diffusion and shielding the cycle from uncontrolled explosive consumption.5 The website must draw a direct analogy between this biological model and the necessity of "second-order constraints" in AI: if an artificial neural network is breached by an environmental disturbance (such as an influx of novel, high-entropy data), the system's internal processes must automatically generate structural responses (computational waste/capsids) to encapsulate the novelty and repair the breach until the system returns to a stable, inert state.5 The site should reference the 2D spatial-temporal Autogenic Automaton particle system simulation to prove that self-preservation can be algorithmically bred through constraint topology.5

Integrating Turney's Model-S and the Immigration Game

To further ground the platform's theoretical strategy, the website updates must integrate Peter Turney's Model-S, a computational framework that abstracts the dynamics of biological evolution within Conway's Game of Life to study symbiogenesis.5 The site must present a clear analogical mapping between biological concepts and their computational execution in artificial environments to explain how teleodynamic AI breeds its structural topologies. The documentation must incorporate the following structural comparisons derived from Model-S to explain evolutionary algorithmic mechanics to site visitors:

The site must emphasize that Turney's experiments, analyzed via apgsearch software, demonstrate that evolutionary pressure does not select for simple, aggressive spatial expansion; instead, fitness in the Immigration Game is positively correlated with the diversity of multicellular autopoietic structures.5 For teleodynamic.com, this translates into a strict architectural requirement: standard incremental parametric mutations are rarely sufficient to drive major leaps in systemic robustness.5 Radical advances in machine intelligence require the synergistic fusion of formerly distinct algorithmic submodels, mimicking the emergence of the eukaryotic cell through symbiogenesis.5

The Expression-Concept Gap and Peircean Semiotics

The website's documentation concerning glyph evidence channels and the ɪ≃1 converter must be thoroughly expanded to address the semiotic nature of teleodynamic intelligence. Teleodynamic architectures mandate a transition in how machine systems represent information, relying heavily on Charles Sanders Peirce’s triadic semiotics—icon, index, and symbol.5 The site must explain that current deep learning is overwhelmingly iconic (mapping similarities in vector spaces) and indexical (identifying statistical correlations), but fails to maintain a self-referential symbolic architecture.5 This semiotic transition necessitates an explicit, highly detailed page dedicated to the "Expression-Concept Gap".8 In standard generative models, fluent token continuation does not equate to a stable concept, leading to hallucinations.8 The website must explain that teleodynamic AI treats this gap as an inspectable design problem, rigidly separating the visible expression from the inferred concept.8 Expressions (Unicode sequences, grapheme clusters, mathematical marks, or glyph phrases like ɪ≃1) contain no intrinsic meaning, while Concepts are the inferred, defended interpretations.8 The site must document the seven-step design model required to cross this gap safely: receiving the expression, normalizing and segmenting it, preserving the surface form, retrieving candidate concepts, checking ontology constraints, emitting a bounded gloss, and storing an immutable trace.8 The core design principle, which must be highlighted prominently on the platform, dictates that a system must never collapse the rendered glyph, the inferred concept, the provenance, and the confidence score into a single registry row.8 In the context of the IOTA-1 approximate public-symbol interpretation system, an input expression such as "智名意" must output a structured record detailing the surface form, the bounded gloss ("AI expression-concept record"), the supporting evidence, and warnings regarding provenance, rather than a direct, unexamined English translation.8 Furthermore, the site must align this with the CLOSET framework (Culture, Language, Organization, Science, Economics, and Technology).5 The website must articulate that symbolic elements are obligate symbionts entirely dependent upon human hosts for cognitive energy, yet they undergo Darwinian evolution and enact self-correcting mechanisms.5 A teleodynamic AI actively treats the acquisition of these semantic structures as the cultivation of an obligate symbiont, using its internal economy to evaluate whether a new symbolic distinction justifies its computational cost, thereby elevating the machine to an oriented semantic system.5

Documentation of the Two-Timescale Reference Architecture

To achieve endogenous constraint closure without collapsing under computational bloat, teleodynamic AI must simultaneously govern continuous parametric adaptation and discrete structural evolution. The teleodynamic.com architecture page must visually and textually detail the Two-Timescale Reference Architecture, creating a clear bifurcation of operational dynamics linked by a synthesized resource variable.2

Inner Dynamics: The Fast Loop

The updated technical documentation must explicitly define the inner dynamics, or the "Fast Loop," which operates continuously at high frequencies.8 The site must explain that this loop processes incoming data samples and executes standard inference upon a temporarily fixed structural capacity. Crucially, the website must update its mathematical specifications to reflect the findings of ter Horst and Zambrano, stating that the fast loop updates parametric weights using natural gradient descent on the parameter manifold.1 The site must display the mathematical formulation of the inner dynamic update step: [Figure omitted from source export] The supporting text must clarify that [Figure omitted from source export] represents the learning rate, [Figure omitted from source export] denotes the inverse Fisher information matrix governing the Riemannian geometry of the parameter space, and [Figure omitted from source export] represents the cross-entropy loss over the input [Figure omitted from source export] and target [Figure omitted from source export].1 The documentation must stress that during the execution of the fast loop, structural actions are assigned an energy cost of zero ([Figure omitted from source export]), allowing the system to fully optimize existing structural constraints before demanding a topological shift.1

Outer Dynamics: The Slow Loop

Conversely, the outer dynamics, or the "Slow Loop," must be documented as operating discretely and sparingly at lower frequencies.1 The website must explain that this loop is explicitly gated by the teleodynamic objective function and is responsible for proposing and executing physical structural modifications to the hypothesis set [Figure omitted from source export].1 The site should format the outer dynamic step function as: [Figure omitted from source export] The supporting text must detail that the structural action [Figure omitted from source export] is selected to minimize the localized cost trajectory, requires accumulated evidence (such as minimum positive/negative examples for a wedge/split action), and demands explicit computational energy expenditures.1

Specifying the Endogenous Resource Economy ([Figure omitted from source export])

The linchpin connecting the inner and outer timescales is the endogenous resource state, [Figure omitted from source export]. The website's "Resource Economy" page must be entirely rewritten to contrast teleodynamic learners with standard deep learning models that expand boundlessly using external hardware budgets.4 The documentation must unequivocally state that a teleodynamic learner must physically pay for its representational structure using a synthesized internal viability budget.4 The resource budget functions as a dynamic reservoir of capital. The website must prominently feature the governing resource law: [Figure omitted from source export] The accompanying explanation must detail how the budget is replenished through predictive success and depleted by natural thermodynamic decay, the immediate computational cost of structural edits, and the continuous maintenance burden of sustaining complex network topologies.4 If the system lacks "resource closure"—meaning [Figure omitted from source export] drops near a pre-defined viability floor—all computationally expensive structural actions must be blocked by the architecture, forcing the system into a structural halt until predictive accuracy within the fast loop restores the energy reservoir.4 Within this economy, the website must define the concept of "software work" across four primary operational burdens that constantly drain [Figure omitted from source export]:

1. Compute: The raw processor cycles, GPU utilization, and hardware energy consumed during forward inference, natural gradient updates, database retrieval, and output validation.8
2. Memory: The physical RAM and logical storage arrays required to house parametric weights, maintain structural dependency graphs, index historical data, and retain append-only audit trace histories.8
3. Review: The overarching human governance burden, encompassing post-hoc interpretation tasks, human comprehension testing, source provenance verification, and safety audits.8
4. Uncertainty: The probabilistic pressure exerted by ambiguous, out-of-distribution data, which triggers costly fallback scenarios, ambiguity reduction protocols, and unresolved case handling routines.8

The Local Objective Function

To further clarify the slow loop's decision-making process, the site must document the Local Objective Function. Instead of executing computationally prohibitive outer-loop Neural Architecture Search (NAS) to find a global optimum across a vast topological landscape, the slow loop operates myopically.5 It selects structural edits that minimize a strictly localized objective score given the immediate environmental pressure and the current resource constraints.8 The site must publish the following candidate score equation used to gate structural action selection: [Figure omitted from source export] The architectural documentation must explain that the system selects the candidate yielding the lowest expected [Figure omitted from source export], strictly subject to the condition that [Figure omitted from source export] holds sufficient capital to pay the declared action cost.8 By internalizing the complexity penalty ([Figure omitted from source export]) and the energy penalty ([Figure omitted from source export]), the system guarantees a bias against frivolous topological expansion.

The Structural Operator Library and the Dominance of the No-Op

The website's "Operator Library" must be updated to serve as a practical reference for self-maintaining AI systems.12 To prioritize inspectable, rigorous structural maintenance over chaotic novelty, the site must detail how the slow loop executes changes using a highly constrained, domain-specific, and fully reversible library of structural operators.8 The following structural operators must be documented on the site, explicitly detailing their triggers, declared costs, and required guardrails:

The site's documentation must dedicate an extensive section to the philosophical and computational dominance of the No-op (No Operation). The no-op is the most critical feature differentiating teleodynamic AI from standard adaptive learning; it represents the system actively refusing unjustified structural growth to conserve computational energy.1 The site must state that choosing a no-op is an explicit, selectable, and affordable action resulting from the calculation of the internal resource economy.8 Without active no-op dominance, the accumulation of new features and semantic distinctions becomes mere uninterpretable "clutter," plaguing massive language models.8 Furthermore, explicit cooldown periods following any structural action heavily enforce this bias, preventing rapid, chaotic structural oscillation.1

Empirical Grounding: Showcasing the Distinction Engine (DE11)

To validate the teleodynamic framework, the website must be updated to prominently feature the empirical findings from ter Horst and Zambrano's 2026 research.1 The theoretical postulations of teleodynamics are concretely instantiated in their system, known as the Distinction Engine (DE11).2 The site must detail how DE11 moves away from conventional continuous-space neural architectures to ground its learning mechanics in George Spencer-Brown’s seminal Laws of Form, utilizing information geometry and tropical optimization (min-plus algebra) to select discrete structural changes endogenously.1 The platform must explain the coalgebraic step function that interleaves probabilistic inference via softmax parameters with the tropical selection of winning structural hypotheses, thereby unifying regularization, architecture search, and resource-bounded inference into a single overarching principle.1 Crucially, the website must display the performance benchmarks validating this approach. The updates must state that on standard classification benchmarks, DE11 achieved 93.3 percent test accuracy on the IRIS dataset (outperforming traditional 91.1% logistic regression baselines), 92.6 percent on the WINE dataset, and 94.7 percent on the Breast Cancer diagnostic benchmark.1 The narrative must emphasize that more profoundly than the accuracy metrics, DE11 achieves these results while generating interpretable logical rules that arise purely endogenously from the coupled inner and outer learning dynamics, rather than being imposed by manual feature engineering.1 The system demonstrates true self-stabilization without any externally imposed stopping rules; after a finite number of outer dynamic structural actions, the architecture autonomously freezes its topology when further additions fail to yield sufficient energy reduction, entering a purely parametric tuning phase.1

Integrating Offline AI and Local Endpoint Sandboxes

A critical missing component in the previous iteration of the website's architecture documentation was the absolute necessity of localized, offline AI execution environments.18 Most contemporary AI demonstrations rely heavily on centralized cloud endpoints, which introduce unacceptable latency, ongoing token accumulation costs, and severe data privacy vulnerabilities.19 For teleodynamic systems—which must continuously evaluate structural viability through physical simulation and constant fast-loop interactions—cloud dependency fundamentally breaks the [Figure omitted from source export] economy.19 Therefore, teleodynamic.com must update its integration guides to extensively detail how localized evaluation infrastructures act as critical, auditable sandboxes.5 The website must provide comprehensive tutorials and architectural diagrams for utilizing Foundry Local, which provides an OpenAI-compatible endpoint running entirely on-device without cloud subscriptions, API keys, or internet requirements.19 Furthermore, the platform must outline the integration of the Microsoft Agent Framework coupled with physics simulators like PyBullet to translate natural language teleodynamic directives into validated, safe physical actions (such as robotic arm manipulation).19 The site's documentation must detail the exact agent orchestration architecture required for these local sandboxes:

The website must emphasize that deploying self-modifying teleodynamic systems via tools like LocalEndpoint, LM Studio, and Pi Agentic Coder guarantees that structural modifications maintain rigorous public-symbol boundaries.5 By running Qwen and Gemma models entirely locally, the structure, cost, viability, and uncertainty of the self-maintaining artificial intelligence remain perfectly auditable and isolated from external API volatility.5

Redesigning the Memory Ecosystems Hub

Memory within teleodynamic architectures is deliberately bifurcated to serve as both an epistemic safeguard and a metabolic relief valve, preventing the catastrophic computational overhead associated with housing all historical data within active parametric weights.4 The website must prominently feature the distinct setups for short-term volatile memory and reviewed, long-term preservation ecosystems.4 The platform must update the UI pathways directing users to the specific utility wizards that govern these memory layers. For short-term AI memory, the site must detail the UAIX AI Memory Package Wizard, instructing users on how to generate compact local handoff files, receiver briefs, startup packets, and optional LLM Wiki plans.4 This localized architecture allows for highly fluid, rapid contextual updates without permanently polluting the deeper semantic constraints of the learner.4 For intermediate organizational structures, the site must guide users to the LLMWikis Setup Wizard, outlining the process for planning source policy, establishing trust labels, configuring wiki structure, establishing governance models, and mapping agent reading paths.4 For permanent, long-term memory, the system must interface directly with decentralized, reviewed repositories. The website must extensively document the integration with AIWikis.org, which is tasked with preserving meticulously reviewed evaluation reports, systemic decisions, source routing data, immutable cryptographic checksums, and public-safe summaries.4 Furthermore, the theoretical documentation must elucidate the role of the NeuralWikis Exchange (neurowikis.com). Originally designed as a collaborative repository for neurology diagnostics, it has been rearchitected into a self-moderated memory exchange to support teleodynamic infrastructure.5 NeuralWikis operates as the intermediary ecosystem orchestrating the interactions between autonomous computational agents, evolving semantic memory structures, and human governance operators.5 To securely translate machine-readable teleodynamic workflows into auditable, human-readable encyclopedic processes, the exchange enforces multi-layered memory firewalls and relies upon a specialized tri-modal consensus protocol.5 The site must state that these firewalls are essential for preventing corrupted parametric states or unresolved high-entropy nodes from propagating into the permanent governance interface.5

Evolutionary Morphodynamics and Agricultural SOMs

The website should briefly contextualize how these memory maps are bred by drawing parallels to biological predictive breeding.5 The documentation should reference how Self-Organizing Maps (SOMs) are utilized in plant science to map non-linear genotypic inputs and environmental covariates.5 By highlighting specific empirical examples—such as evaluating spatiotemporal germplasm trajectories in potato lineages (Solanum tuberosum), assessing genomic values in maize (Zea mays L.) under temperature stress, unraveling cold defense mechanisms in grapevine transcriptomics, and guiding diallel crosses in pepper (Capsicum annum) and rice (Oriza sativa L.)—the site grounds the abstract AI concepts in proven evolutionary methodologies.5 Teleodynamic AI synthesizes these principles, treating SOM-generated neural clusters as proposed phenotypic traits, evaluating whether the newly bred representational cluster provides enough predictive utility to justify its placement in long-term memory.5

The Evaluation Lab: Metric Families and Review Gates

The website's "Evaluation Lab" section must be thoroughly expanded to clarify that standard deep learning benchmark scores alone are entirely insufficient to prove teleodynamic behavior.8 Teleodynamic AI must be evaluated beyond mere accuracy, focusing intensively on viability, structural history, compatibility, stability, interpretability, and human comprehension.12 The site must publish comprehensive documentation regarding the expanded Metric Families utilized to assess the architecture:

Furthermore, the website must rigidly document the Acceptance Gates (Review Gates) that protect public claim boundaries. Every proposed structural modification or semantic output must sequentially pass specific conditions; failure triggers an automated, immutable response protocol.8

The Evaluation Lab portal must also detail the required Audit Trails and Phase Publications.8 To verify a teleodynamic claim, the system must export fully reconstructable audit packages utilizing Glyph Review Records (Review Worksheets).8 The website must list the required fields for these worksheets: input sequence, original source, normalized forms, grapheme clusters, private-use checks, candidate glosses, real-time [Figure omitted from source export] status, the exact structural action taken, human comprehension results, and the final classification status (Draft, Emerging, Stable, or Rejected).8 Finally, the UI must provide dashboard examples for mandated evaluation panels. These must include the Stability Plot (tracking structural actions per 1,000 inputs to demonstrate discovery rises and stabilization plateaus), the Pareto Front (visualizing the multi-objective tradeoff between accuracy, structural complexity, and consumed energy), and Viability Retention tracking (ensuring [Figure omitted from source export] remains above the viability floor under adversarial input streams).8

Staged Implementation Roadmap Updates

To guide engineers visiting the site, the "Roadmap" section must clearly define the rigid progression sequence required to build a production-ready teleodynamic AI. The site must declare the absolute definition of done: a prototype cannot be deemed teleodynamic until its split, merge, add, retire, and no-op decisions are completely resource-gated, locally justified, appended to immutable logs, and distinctly visible within phase plots.12 The documentation must strictly warn developers against beginning with massive, heavily structured parameter stores. Instead, the implementation must follow a systematically staged, six-phase build sequence that layers viability states over a deliberately minimal substrate.12

Through these sweeping, highly granular updates, teleodynamic.com will successfully pivot from a speculative architectural catalog into the definitive, empirical authority on resource-bounded, constraint-maintaining artificial intelligence. By intertwining deep theoretical exposition—spanning Deacon's hierarchy and Peirce's semiotics—with hard empirical benchmarks from the DE11 Distinction Engine and rigorous architectural specifications for local endpoint sandboxing, the platform will deliver a peerless public conceptual experience tailored to the vanguard of the AI engineering community.

Works cited

1. Teleodynamic Learning a new Paradigm For Interpretable AI \- arXiv, accessed June 2, 2026, https://arxiv.org/pdf/2603.11355
2. \[2603.11355\] Teleodynamic Learning a new Paradigm For Interpretable AI \- arXiv, accessed June 2, 2026, https://arxiv.org/abs/2603.11355
3. Research Foundations for Teleodynamic AI \- Teleodynamic.com, accessed June 2, 2026, https://teleodynamic.com/research-foundations/
4. Teleodynamic AI, accessed June 2, 2026, https://teleodynamic.com/
5. Developing Teleodynamic AI with LocalEndpoint.md
6. Teleodynamic Learning a new Paradigm For Interpretable AI | Request PDF \- ResearchGate, accessed June 2, 2026, https://www.researchgate.net/publication/401909814\_Teleodynamic\_Learning\_a\_new\_Paradigm\_For\_Interpretable\_AI
7. Enrique TER HORST | Associate Professor | PhD | Los Andes University (Colombia), Bogotá | UNIANDES | Faculty of Administration | Research profile \- ResearchGate, accessed June 2, 2026, https://www.researchgate.net/profile/Enrique-Ter-Horst
8. Start Here: Teleodynamic AI in Plain Terms \- Teleodynamic.com, accessed June 2, 2026, https://teleodynamic.com/start-here/
9. Contact Michael Kappel | Teleodynamic.com, accessed June 2, 2026, https://teleodynamic.com/contact/
10. Incomplete nature: how mind emerged from matter | Request PDF \- ResearchGate, accessed June 2, 2026, https://www.researchgate.net/publication/320181769\_Incomplete\_nature\_how\_mind\_emerged\_from\_matter
11. Teleodynamic AI Glossary, accessed June 2, 2026, https://teleodynamic.com/glossary/
12. Teleodynamic AI Roadmap for Self-Maintaining Systems, accessed June 2, 2026, https://teleodynamic.com/roadmap/
13. Evolution of Symbiosis in the Game of Life: Three Characteristics of Successful Symbiotes \- arXiv, accessed June 2, 2026, https://arxiv.org/pdf/2104.01242
14. Evolution of Autopoiesis and Multicellularity in the Game of Life \- MIT Press Direct, accessed June 2, 2026, https://direct.mit.edu/artl/article/27/1/26/101060/Evolution-of-Autopoiesis-and-Multicellularity-in
15. (PDF) Evolution, Symbiosis, and Autopoiesis in the Game of Life \- ResearchGate, accessed June 2, 2026, https://www.researchgate.net/publication/344373576\_Evolution\_Symbiosis\_and\_Autopoiesis\_in\_the\_Game\_of\_Life
16. Symbiosis Promotes Fitness Improvements in the Game of Life \- MIT Press Direct, accessed June 2, 2026, https://direct.mit.edu/artl/article/26/3/338/93245/Symbiosis-Promotes-Fitness-Improvements-in-the
17. Evolution of Autopoiesis and Multicellularity in the Game of Life \- PubMed, accessed June 2, 2026, https://pubmed.ncbi.nlm.nih.gov/34529755/
18. The Best Local Agentic Coding Workflow (Complete Guide) \- YouTube, accessed June 2, 2026, https://www.youtube.com/watch?v=UngVdAsQEiU
19. Building real-world AI automation with Foundry Local and the Microsoft Agent Framework, accessed June 2, 2026, https://techcommunity.microsoft.com/blog/azuredevcommunityblog/building-real-world-ai-automation-with-foundry-local-and-the-microsoft-agent-fra/4501898
20. Running LLMs Locally – Tether's Paolo Ardoino Reveals Decentralized AI Platform QVAC \- YouTube, accessed June 2, 2026, https://www.youtube.com/watch?v=8ZyzH24vaQE