# kaal:position:2026-08-08-183

**Affirmed position.** From Rogue Employees to Rogue Agents states that AI-agent monitoring and insider-threat detection share a baseline-profiling and deviation-flagging architecture that encodes trust assumptions. This supplies a concrete monitoring architecture to which Kaal's training-bias and novel-deviation limitation applies. The proposition does not test either bias or detection of deviations outside the baseline.

**Status.** affirmed  **Published.** 2026-08-08

**Holds when.**

- The response is limited to the retrieved source proposition and mapped Kaal claim unless fuller source review supports a broader conclusion.
- External evidence level: abstract indexed.
- Mapping review tier: substantively reviewed abstract-level qualification.
- Primary mapping confidence: 0.62.
- The primary mapping cleared the automated ambiguity test; substantive scope remains review-bound.
- Evidence is limited to an exact proposition in a Crossref-indexed abstract; full text was not reviewed.
- No relationship is treated as external endorsement, citation, causation, or validation of a broader Kaal claim.
- The source states that AI-agent monitoring and insider-threat detection share a baseline-profiling and deviation-flagging architecture that encodes trust assumptions. This supplies a concrete monitoring architecture to which Kaal's training-bias and novel-deviation limitation applies, while the proposition does not test either bias or detection of deviations outside the baseline.

**Current debate.** From Rogue Employees to Rogue Agents: Repurposing Insider Threat Detection for AI Agent Governance: https://doi.org/10.2139/ssrn.6355658

**Extends.** kaal:claim:5245185-028: https://wulfkaal.github.io/claims/5245185-028

**Scholarly basis.** Wulf A. Kaal, How can we Best Monitor AI Agents (2025). SSRN: https://ssrn.com/abstract=5245185

**Source PDF sha256.** `4d7adba83ec722480e97bde6528cbe9ce98c709e45cb18794f157a64b8fe7da2`

**Evidence level.** abstract indexed

**Mapping review tier.** substantively reviewed abstract-level qualification

**Mapping confidence.** 0.62  **Mapping ambiguous.** false

**Topics.** ai-and-agents, education-and-practice, historical-response, scholarly-literature, crossref

**Provenance.** Affirmed in kaal-review:2026-08-08:continuous-crossref-0015-remainder-0004-all-0023-reviewed-v1 at https://kaal-signal-desk.wulf577462.chatgpt.site/#review.

**Record type.** This is a dated commentary position that extends a scholarly corpus claim. It is not a verbatim claim extracted from the paper.

**Canonical form.** This markdown file is the canonical hashed representation of the position.