An Evidence Model for Agentic Processes: Evidence Claims, Trust Assumptions, and Policy Assessment

Abstract

Agentic AI systems increasingly exchange messages, invoke tools, request approvals, hold structured decision sessions, and modify shared artifacts. Logs and anchors can make selected records tamper-evident, but they can also mislead if their evidentiary meaning is implicit: a hash does not establish semantic truth, a signature does not establish authorization, and an external anchor does not establish capture completeness. This paper proposes an evidence claim model for agentic processes. It dis

Categories

@misc{bromme2026evidence,
  title = {{An Evidence Model for Agentic Processes: Evidence Claims, Trust Assumptions, and Policy Assessment}},
  author = {Arslan Brömme},
  year = {2026},
  month = sep,
  eprint = {2609.08481},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.08481}
}