Workstream 01
Matched ecological controls and separate history/substrate interventions
AI Labs FutureWorks Research profile
A controlled emergence ecology for testing persistent identity, history, diverse substrates, and governance needs.
A Compacitas research initiative
AletheonAgent behavior
Aletheon investigates whether a division of governance work can develop without assigning archetypal roles in advance. Its current protocol separates persistent identity and history from the reasoning substrate and tests both against shuffled, homogeneous, rotating-identity, and neutral controls. The proposed experiment combines brief, receipted model-learning windows with a restricted executable policy capsule. It evaluates whether any specialization depends on a real ecological need, transfers to new conditions, and can reconstitute after disruptions.
The research question
Research workstreams
Matched ecological controls and separate history/substrate interventions
Restricted policy capsules and causal execution evidence
Transfer, disruption, reconstitution, and concentration tests
Evidence priorities
The initiative is framed around testable questions. Its research direction should be assessed against observed evidence and the limits of each experiment.
Research goals and proposed mechanisms are not established findings. Product-linked tracks describe the research behind a capability, rather than a claim that its hypotheses have been validated.
Associated company
The verifiable civilization runtime for model governance, scheduling, optimization, and authorized compute federation.
Explore Compacitas
MultiplexCompacitas · Prototype
Scheduling many persistent agent identities across bounded model residency and physical compute.
Verifiable Civilizational LearningCompacitas · Proposed study
Testing whether verified experience can accumulate outside model weights and survive complete agent and model turnover.
A System of Heredity for IntelligenceCompacitas · Research design
Checked inheritance of commitments and capability across generations of neural and non-neural computational systems.