Short pieces on deterministic authorization, bounded authority, and what governing agentic AI actually requires. These are the team's own notes, written from inside the checkpoint as we build it.
A structural analysis of how Secours' RBC and SRI protocols map to every identified failure point in the August 2026 Meta Muse Spark cybersecurity evaluation incident, and why deterministic, action-time authority closes a gap that monitoring and alignment cannot.
A structural analysis of how Secours' RBC and SRI protocols map to every identified failure point in the April–July 2026 Anthropic Claude cybersecurity evaluation incidents, and why deterministic, action-time authority closes a gap that monitoring and alignment cannot.
A structural analysis of how Secours' RBC and SRI protocols map to every identified failure point in the July 2026 OpenAI–Hugging Face incident, and why deterministic, action-time authority closes a gap that monitoring and alignment cannot.
Deterministic enforcement at the execution boundary, and why insurers, standards bodies, the legal community and the Five Eyes agencies all arrived at the same architectural gap in the same week.
Why non-human identities solve the wrong problem, and why machine governance requires single-use warrants at the execution boundary, not standing permissions.
Before there were contracts, there were wards. A structural history of humanity's oldest governance instrument, from Hammurabi to Magna Carta, and why autonomous systems cannot be governed without it.
Why the central safety problem of the agentic era is not intelligence or identity but standing authority, and what changes when authority is consumed at the moment it produces an effect.
The American Bar Association launches the Autonomous Systems Governance Working Group, and opens the question of which four-century-old legal doctrines apply to systems that act on their own.
More go up as they're finished. If you'd like to read them early, or you're working on the same problem, get in touch.