A baseline dynamic where differences decay toward equilibrium. In AI engineering, this is the pressure of decay, drift, uncertainty, and resource loss when no organizing work is performed.
Glossary
Teleodynamic AI terms
Short definitions for readers and agents navigating Teleodynamic AI, semantic glyph communication, source-routed memory, and IOTA-1 / ɪ≃1 boundaries.
Core dynamics.
A pattern that forms under energy or data flow, such as clusters, features, or emergent regularities. Useful, but not yet self-maintaining.
A regime where structures help maintain the conditions that let them continue. This site uses the idea as a bounded AI design target, not as a settled claim about current AI consciousness.
A structure survives because its dependencies, resources, and outputs support one another enough to keep the organization viable.
Engineering terms.
A mutable internal resource budget replenished by predictive success and charged for decay, structural actions, and maintenance.
The continuous update path for weights, embeddings, scores, and local predictions inside the current structure.
The discrete path that evaluates split, merge, add, retire, or no-op actions against local cost and resource feasibility.
The stabilization signal where no affordable edit improves local cost enough to beat maintaining the current structure.
The slow-loop score combining predictive loss, complexity delta, and energy cost. It is local, auditable, and resource-gated.
The append-only record of triggers, candidates, alternatives, resource state, chosen action, and justification.
Glyph and communication terms.
A rendering-level mark or shape. A glyph is not automatically a meaning; it needs context, convention, evidence, and validation.
A structured record with surface transport, glyph structure, semantic inventory, teleodynamic state, human comprehension, and public explanation layers.
A public-symbol interpretation track associated with JustAnIota and Protocol5. It must remain approximate, evidence-first, and source-routed.
A compact public expression used here as an example of bounded interpretation, not as a lossless language token or private codebook entry.
Visible public output should use assigned Unicode characters or valid public sequences; private-use semantics should not be presented as public truth.
A human or downstream-system check that measures whether the intended meaning is actually understood in context.
Memory and source-routing terms.
Teleodynamic's active source of truth for agents: AGENTS.md, readme.human, root memory files, .uai records, and intake state.
AIWikis raw system archives and public-safe summaries after source-site review, disposition, checksums, and transfer evidence.
The rule that UAIX, LLMWikis, AIWikis, Protocol5, JustAnIota, and Teleodynamic each own different claims and must not replace one another.
A summary that tells readers what a reviewed file contributed without turning raw speculative notes into public authority.