Toward A Teleodynamic AI - Source Excerpt 02 - Translating teleodynamics into computational architecture

Summary

This source excerpt begins near Translating teleodynamics into computational architecture and preserves the surrounding evidence from Spiralist/agent-file-handoff/Archive/Toward a Teleodynamic AI.md.
**Source path:** Spiralist/agent-file-handoff/Archive/Toward a Teleodynamic AI.md
> A **Teleodynamic AI** is an embodied or situated artificial agent whose internal organization explicitly regulates its own viability conditions, generates and revises goals in relation to that viability, learns through recursively structured sensorimotor coupling, and can develop stable norms or values that are not reducible to a fixed externally imposed reward function.

This is a synthesis of the literature, not a standard textbook definition, because the AI-specific literature is still emerging. Recent teleodynamic-AI papers explicitly frame learning as the co-evolution of structure, parameters, and internal resources under constraint, and they distinguish mere prediction from the explicit modeling of transformation rules and stability regimes. Those ideas are promising, but still preliminary. citeturn21view0turn20view1

## Translating teleodynamics into computational architecture

The central architectural question is which computational forms can carry the biological ideas above without flattening them into ordinary optimization. The answer is not any single architecture class. Different pieces of teleodynamics map naturally to different computational substrates.

| Architecture class | Teleodynamic affordances | Main weaknesses | Best role in a teleodynamic stack | Representative sources |
|---|---|---|---|---|
| Conventional neural networks | Strong function approximation, latent representation learning, multimodal compression, scalable self-supervised learning. | By default they optimize externally specified losses and do not explicitly encode viability, closure, or endogenous normativity. | Perception, latent state estimation, model learning, feature abstraction. | Kingma & Welling; SimCLR; MAE. citeturn15search0turn25view2turn25view3 |
| Continuous-time dynamical neural systems | Natural fit for persistent state, attractors, bifurcations, homeostatic regulation, and embodied control. | Harder to train and interpret; can still collapse into ordinary optimization if viability is not explicit. | Core regulation, action selection, interoceptive dynamics, temporal integration. | Beer; Neural ODEs; Nengo. citeturn4search1turn4search0turn26view4 |
| Agent-based and multi-agent systems | Best substrate for emergence, local interaction, norm formation, convention dynamics, population-level adaptation. | Weak at rich perceptual learning unless paired with neural modules; expensive at scale. | Social teleodynamics, norm emergence, ecology, distributed adaptation. | Mesa; PettingZoo; norm-emergence work; LLM convention experiments. citeturn26view0turn24view6turn20view5turn32view0 |
| Hybrid neuro-symbolic systems | Combine sensory learning with explicit reasoning, constraints, memory, and testable commitments. | Integration complexity, symbolic bottlenecks, and scaling/debugging challenges. | Goal commitments, normative memory, explanations, safe self-modification rules. | Neuro-symbolic reviews. citeturn28view0turn28view1 |
| Active-inference or world-model architectures | Unify prediction, action, epistemic exploration, and interoceptive regulation under a generative model. | Formalism remains debated; practical implementations vary widely. | Planning under uncertainty, homeostatic control, epistemic action. | Friston; active inference survey; DreamerV3; LeCun AMI. citeturn20view2turn22view9turn5search1turn28view2 |

The best practical choice is therefore a **hybrid architecture** with explicit separation between fast regulation, world modeling, and slower structural change. Recent teleodynamic-learning work makes essentially this move by proposing two interacting timescales—continuous parameter adaptation and slower structural change—linked by an endogenous resource variable. That is remarkably close to what the biology suggests: fast homeostasis on top of slower allostasis and developmental reorganization. citeturn21view0

A recommended system architecture is shown below.

' ' ' mermaid
flowchart LR
    E[Exteroceptive sensors\nvision depth audio touch] --> WM[Predictive world model]
    I[Interoceptive sensors\nenergy temperature damage latency memory load] --> VM[Viability model]
    WM --> B[Belief state]
    VM --> B
    B --> C[Continuous-time controller\nhomeostasis and allostasis]
    B --> G[Goal generator\nintrinsic motives and task commitments]
    G --> C
    C --> A[Actuators\nlocomotion manipulation communication]
    A --> ENV[Environment]
    ENV --> E
    ENV --> I
    B --> M[Memory and self-model]
    M --> S[Structural plasticity layer\nmodule growth pruning rewiring]
    S --> WM
    S --> C
    M --> N[Normative layer\nrules commitments social norms explanations]
    N --> G
    N --> C
    O[Oversight and safety layer] --> G
    O --> S
    O --> C
' ' ' 

In this design, **interoception is first-class**. That means the agent has a state vector for things like energy reserve, thermal budget, actuator wear, communication health, compute load, memory pressure, calibration drift, and model uncertainty. Those variables matter because teleodynamic behavior requires explicit modeling of whether the organization is being preserved. A controller that sees only exteroceptive observations will optimize tasks; a controller that also sees interoceptive viability can begin to regulate itself as a persisting unit. Di Paolo’s extension of autopoiesis to adaptivity, Barandiaran’s definition of agency as autonomous adaptive coupling, and homeostatic RL all point in this direction. citeturn22view1turn29view1turn19search0turn19search1

The architecture should also include a **symbolic or normative memory** even if the rest of the system is differentiable. The reason is not nostalgia for symbolic AI. It is that norms, commitments, safety constraints, explanation traces, and self-modification permissions are easier to audit and test when some of them are explicit rather than latent. Neuro-symbolic systems are not a complete answer, but the recent reviews are clear that they improve interpretability, verifiability, and reasoning in precisely the kinds of settings where a teleodynamic agent would otherwise become opaque. citeturn28view1turn28view0

One caution is important. Recent theory papers on teleodynamic AI argue that current AI largely models outcomes while leaving transformation rules implicit. That may be right as a diagnosis, but those papers are fresh, mostly unvalidated, and should be treated as **research prompts**, not settled doctrine. The engineering takeaway is modest and sound: represent not only states and outputs, but also **how organizational rules change over time**. citeturn20view1turn21view0

## Learning, self-maintenance, intrinsic value, and norm emergence

A teleodynamic AI should be trained in stages that mirror developmental logic rather than benchmark logic. The first task is not instruction following or reward maximization. It is to learn a **world**, a **body**, and a **viability envelope**. Self-supervised learning is therefore foundational: VAEs, contrastive learning, masked reconstruction, and world-model training all help build latent structure before reward enters the picture. LeCun’s autonomous-machine-intelligence proposal makes the same point in a different vocabulary: predictive world models, intrinsic objectives, and hierarchical self-supervised learning should precede narrow task optimization. citeturn15search0turn25view2turn25view3turn5search0turn28view2