One world, many entities — and trust that verifies itself
The v0.10 generation of the Camile World Model sustains multiple entities over one world, exchanges evidence with proof of integrity, and makes shared reading auditable. What changes for those building with the CWM.
Until now, the Camile World Model sustained one entity over its world: one identity, one trajectory, one calibration. Version v0.10.3 changes that limit — the model now accommodates multiple digital entities simultaneously over the same world state, each with its own view of itself and its own history of confidence.
The critical problem solved is isolation: one entity's experience does not contaminate the calibration of the others. Every identity keeps its own gauge of how reliable it has been — and each entity's birth is recorded in an auditable way, with its genealogy preserved.
The model now registers and serves multiple entities over the same world, each with its own self-model: how it sees itself within the shared world.
Confidence calibration became isolated per entity — same mathematics, separate instances; one entity's performance does not change the confidence of others.
Each entity's birth and identity are recorded deterministically and auditably, ensuring any party can reconstruct who is who.
Every synchronization message now includes a proof of integrity that the receiver re-executes locally — verification is a computation, not an act of faith.
Disagreements between observers gained first-class treatment: instead of a merge that hides the conflict, the system issues a receipt with both versions and their attributions.
A cross-grading mechanic exposes claims that fail when measured against the declarant's own real calibration.
Multi-agent systems can collaborate on shared decisions — joint forecasts, alerts, chained evidence — without any participant blindly trusting the others.
Integrity audits no longer require system access: whoever received a message can prove it is intact and who sent it, using only their own data.
Multi-tenant platforms and joint operations gain a shared world state with read isolation and open audit — without duplicating data and without blind trust spots.
Transparency commitments stop depending on manual reports: the honesty indicator is computed and published continuously.
A shared world is only as trustworthy as the ability to check it. v0.10.5 delivers that ability as part of the contract.