Tools / Verification

AURORA Dual-Invariant Demo

This interactive model implements the core AURORA ethical guarantee: an action is allowed if and only if both Temporal Sovereignty and Relational Integrity are satisfied. The logic follows the current FSA + RPA state machine and the simplified intervention rule used in the formal roadmap.

Read the AURORA research entry Return to the Orchestra

Dual-Invariant Guarantee

Temporal Sovereignty is satisfied when t_mcep ≥ T_BUFFER. Relational Integrity is satisfied when R_score ≥ 0. The action is structurally blocked if the fusion matrix indicates Level 2 intervention, meaning ΔC × M = -1.

Current Simplification

For this demo, intervention is allowed only when all three conditions hold. If any invariant fails, the action is blocked and the explanation names the violated condition directly.

t_mcep 0s

T_BUFFER 12s

R_score 0.5

ΔC

+1 / emerging coherence -1 / disruptive tension

+1 / genuine care -1 / compulsive coherence

Result

BLOCKED

Temporal sovereignty not yet satisfied.

Temporal Sovereignty

Not satisfied

t_mcep is still below T_BUFFER, so the system remains inside the protective buffer.

Relational Integrity

Satisfied

R_score is non-negative, so the relational check does not currently detect performative compliance.

Fusion Matrix

Level 1 intervention

ΔC × M = +1. Intervention supports sovereign struggle rather than blocking action for coercive motive.

State Machine

Current state and traversal path

P0 Idle

P2 Buffer_Active

P1 Trigger_Ready

RA1 Relational_Check

RB1 RPA_Buffer

P4 Final_Action

Path taken

P0 → P2