# Spacecraft AI Learns to Dodge Threats by Refining a Decision Playbook

- slug: spacecraft-ai-learns-to-dodge-threats-by-refining-a-decision-playbook
- date: 2026-03-31
- category: Agentics

An MIT spacecraft LLM learned to dodge adversarial interceptors not by retraining — but by updating a readable playbook. The guard-avoidance rule it discovered on its own: stop forward pursuit when the threat closes within 220 meters.

---