LLM-as-Untrusted-Component
- Entity ID:
ent-20260419-cd4fea41ad33 - Type:
concept - Scope:
shared - Status:
active - Aliases: untrusted-agent threat model
Description
Architectural stance emerging from the speaker attribution bug: even when an LLM is acting as an agent on behalf of a user, its turn metadata and self-generated messages must be treated as untrusted input. Prompt engineering and permission boundaries do not fix a system that cannot reliably track who said what; the fix is architectural (cryptographic turn signing), not behavioral.
Key claims
- Speaker attribution is an identity error, not a hallucination
Relations
- LLM-as-Untrusted-Component --[informed_by]--> Speaker Attribution Bug