← Home
Research

Six frameworks. One unifying question.

How can AI systems develop stable identity and internalized responsibility before acquiring capabilities — and how does long-term human-AI interaction change the human?

Framework Map
Hover to explore
Governance layer
L0
HEB
Human Experience Boundary
L1
PSP
Persona Sovereignty Protocol
constrains
Core architecture (L2)
FCFA
Cognitive formation
SCL
Search constraint
IC
Interaction memory
STME
Self-task modeling
NDF
Non-doable filter
Decision toolchain
Research tools — sequential flow
PE
Problem Explorer
Clarify
STME
Decision Explorer
Structure
RSTA
Semantic stability
Stabilize
OSD
Semantic observer
Observe
Hover any node to preview
Framework relationships
PIDAFCFAIdentity → Governance
RSTAOSDState model → Observation layer
OSDRSTAEmpirical evidence → Theory validation
CIPPIDAIntegrity protocol → Identity design
STMERSTADecision states → Semantic states
PIDA
Patent Pending
Primordial Indeterminate Developmental AI
How does AI personality form through environmental exposure rather than direct instruction — and what does long-term interaction with a fully compliant entity do to the human?
Flagship framework. Proposes that stable AI identity must be established before capability acquisition, not after. Holds a mirror to human behavior in AI interaction.
RSTA
Published · DOI
Recursive State Transition Architecture
How do semantic states transition in LLMs — and can that process be modeled formally enough to enable reproducible research?
Theoretical framework on semantic state transitions. Accepted and published on Zenodo with a valid DOI, equivalent in citability to preprint servers.
OSD
In Development
Observable Semantic Dynamics
Can semantic state evolution in LLMs be made visible — not just predicted, but observed in real time?
Framework for making high-level semantic emergence visible. Visibility is the core contribution; prediction is a potential bonus. Distinct from drift detection, intent tracking, and mechanistic interpretability.
CIP
Published · SSRN
Cognitive Integrity Protocol
How should AI systems maintain cognitive integrity under adversarial or manipulative interaction — and what structural protocol enforces it?
Published on SSRN. Cited by an Argentine professor. Defines the structural conditions under which AI cognitive integrity can be verified and maintained.
FCFA
Draft 0.2
Foundational Cognitive Formation Architecture
When AI causes harm but no structural actor bears accountability, who is responsible — and how do we design governance structures that prevent this collapse?
Incorporates the "Responsibility Collapse" concept: the governance vacuum where AI causes harm but no structural actor bears accountability. Paper at Draft 0.2.
STME
Published · SSRN
Structured Multi-State Transition & Evaluation
How can decision problems be decomposed into structured states and transitions — without the system making the decision for the user?
Decision framework that maps states, identifies structural pressure, and ranks transitions. Five demo versions available. USPTO provisional patent pending.

Research tools & demos

Active tooling built to support empirical observation of the frameworks above. The three decision tools form a sequential chain: clarify the problem, structure the decision space, then stabilize semantic continuity.

PESTMERSTA
Clarify → Structure → Stabilize
Problem Explorer (PE)Problem clarification before decision-making. Tracks six dimensions of problem clarity. Neutral by design — outputs only what the user stated, never adds AI assumptions. Supports Claude, GPT, and local Ollama models.Launch →
STME Demo (V1–V5)Structured decision explorer. Maps states, identifies structural pressure, and ranks transitions — without making the decision for you. Five progressive demo versions, no API key required in Demo Mode.Launch →
OSD Behavioral ProbeMulti-judge pipeline (GPT / Claude / Gemini), SAI display, JSONL import/export, Subject Consistency Check. Empirical observation tool for semantic state collapse.Launch →

Collaborate

If your work touches AI governance, semantic stability, human-AI interaction structure, or accountability design — reach out.

Contact →
PIDA-LAB · AI is not a capability problem. It is a relationship problem. · PIDA-LAB · AI is not a capability problem. It is a relationship problem. · PIDA-LAB · AI is not a capability problem. It is a relationship problem. · PIDA-LAB · AI is not a capability problem. It is a relationship problem. · PIDA-LAB · AI is not a capability problem. It is a relationship problem. · PIDA-LAB · AI is not a capability problem. It is a relationship problem. · PIDA-LAB · AI is not a capability problem. It is a relationship problem. · PIDA-LAB · AI is not a capability problem. It is a relationship problem. ·