What question did this study set out to answer?

The study examines the risks posed by credentialled AI agents compared to human insiders.

June 15, 2026Open Access

The Credentialled Agent Is the New Insider Threat

Key Points

The study examines the risks posed by credentialled AI agents compared to human insiders.
Surveyed enterprise leaders on insider threat perceptions.
Evaluated existing policy frameworks for their adequacy in managing AI agents.
Outlined engineered solutions for governance and accountability.
87% of leaders perceive credentialled AI agents as a significant insider threat.
Traditional policies are inadequate to manage AI agents acting in milliseconds.
Proposed solutions include signed action lineage and a reliable kill-switch.

Abstract

Eighty-seven percent of enterprise leaders now rate a credentialled AI agent a greater insider-threat risk than a human. A look at why policy written for human pace cannot govern a process that acts in milliseconds, and what an engineered answer (signed action lineage, authority-at-execution, a kill-switch that severs authority) actually looks like.Originally published at https://mickai.co.uk/articles/credentialled-agent-new-insider-threat-2026. Mickai is a Sovereign Intelligence Operating System; the Open Audit Record signs every artificial intelligence action before it executes, post-quantum and offline-verifiable. 101 filed UK patent applications, Mickai LTD (Companies House 17166618).

Read Full Paperexternally

Ask AI

Helpful

Bookmark

View Full Paper