Research

Notes and letters on what AI agents do, and what can be checked afterwards. We publish the method and the limits with each one.

• Letter · 14 August 2026Defence in the agent economyAustralia can be exposed to agent-led attacks before its institutions have adopted comparable defensive capability. Model access alone will not close that gap.
• Research note · 3 March 2026An agent can write to its own recordEvery convention for trusting an operational record assumes the actor and the record are separate systems. Agents are the first actors with hands inside the record.
• Research note · 12 November 2025The compounding error problem in production AIWhy benchmark reliability and production reliability are different numbers, the seven conditions that separate them, and why the production figure has to be measured in production.