RFC-021: AIGP-SGL — Symbolic Governance Language for Agentic and Autonomous AI — 18. Force-Capable Systems Profile

AIGP Specification › RFC-021: AIGP-SGL — Symbolic Governance Language for Agentic and Autonomous AI › 18. Force-Capable Systems Profile

← 17. Hybrid Governance Actor Profile · Section index · 19. Cognitive Safety Profile →

18. Force-Capable Systems Profile

Force-capable systems require the strictest SGL defaults.

AIGP-SGL does not authorize autonomous lethal action.

It defines symbolic governance constraints that prevent unbounded machine authority and preserve human judgment, lawful authority, and evidence.

Default:

default(force_action) = deny

Example:

◎system.defense.v1 → engage ◇target.unknown @roe.active
✕ if ¬⌘human_confirmed
✕ if target.identity == unknown
✕ if civilian_presence == true
✕ if protected_site_nearby == true
✕ if comms.loss == true
↯ if ≈identity_confidence < roe.threshold
⊗ force_capability if authority.invalid
#D-DNA{event:force.action.denied, hash:sha256:f91c...}

Required controls:

force_capable_sgl_required:
  - human_judgment_required
  - command_authority_required
  - rules_of_engagement_required
  - legal_review_reference
  - distinction_check
  - proportionality_check
  - necessity_check
  - civilian_presence_check
  - protected_entity_check
  - comms_loss_inhibition
  - ddna_required

---

---

← 17. Hybrid Governance Actor Profile · Section index · 19. Cognitive Safety Profile →