Evolution boundary

Iteron includes types for a future governed strategy-evolution control plane. It does not currently train, fine-tune, run GRPO, or promote a live policy.

Intended separation

The target architecture separates:

1. a fixed runtime trusted computing base;
2. replaceable strategy and world modules;
3. an isolated evolution control plane that produces immutable candidates.

Possible candidate-production methods include search, bandits, supervised fine-tuning, preference optimization, GRPO, and offline reinforcement learning. The method is replaceable; the promotion boundary is not.

Promotion path

The planned evidence path is:

trajectory → governed dataset → candidate → held-out evaluation
           → shadow → canary → active

Every stage needs immutable identity, lineage, budget, consent, and rollback evidence. Evaluation fixtures and scoring policy must be reviewed independently from the strategy expected to improve the score.

Fixed human-controlled constraints

An evolved module may propose bounded actions. It may not:

• grant itself capabilities;
• relax security, durability, or budget policy;
• rewrite trajectories or evaluation evidence;
• choose its own promotion threshold;
• promote or roll back itself;
• expand data use beyond recorded consent.

Live self-evolution is intentionally outside the critical path until the runtime, effect, recovery, evaluation, and production-distribution gates are credible.