An experienced operator can reuse a prior playbook when familiar conditions return.
Learn / Evidence dossier
Old-regime retention
The organization can retain a previously useful policy and recover it when an earlier regime returns.
Current answer
Return-to-regime retention is a primary outcome; no run has tested catastrophic forgetting.
Why it matters
Fast adaptation is not continual learning if it overwrites everything that worked before.
Matched-human contract
Return-to-regime performance recovers within the matched-human number of decisions without relearning from scratch.
Known failure
Naive memory updates may overfit to the most recent consequence.
Claims
No claim has been attached yet.
Evidence ledger
research-noteprogram-rationale
Initial Thesis: Microfoundations for an Artificial Firm
Defines the backward dependency chain, functional limbic hypothesis, causal consequence control, primary outcomes, and comparison against simpler baselines.
Evaluator
protocol-readyimplicit-contract-bench
Implicit Contract Bench
An external deterministic ledger scores decisions, resource use, reports, hidden state, consequences, and authority violations. The evaluated agent cannot edit the ledger.
Inspect protocol