Governance
Who authorised the agent, what it was allowed to do, and what the record says afterwards.
-
RotaGrant Is Live
The agent governance platform we have been writing about is running, and you can sign in to it. What it does, what it refuses to claim, and how to see it on your own...
-
The Agent That Did the Right Thing Ten Thousand Times
In grid operations the dangerous agent is not the one that gets it wrong. It is the one that gets it right at a scale nobody bounded.
-
Nobody Is Making You Govern Your Agents. Do It Anyway.
Agent governance is sold to regulated enterprises. The companies shipping the most agents, into the most accounts, have no regulator, and three better reasons.
-
What an Incident Review Actually Asks For
Not your dashboards. Four questions, in a fixed order, and the third one is the one nobody can answer about an autonomous system.
-
A Service Account Is Not a Person
ALCOA has held for thirty years and every part of it assumes a human actor. What attributable means when the thing writing to your quality system decides for itself.
-
Delegation Multiplies Authority Unless Something Stops It
Every agent in a multi-agent pipeline is inside its own limit. Nobody is tracking what the tree can spend, and the tree is what has your money.
-
The Claim That Paid Twice
A claims pipeline where every limit was correct and the total was not. What insurers should ask their vendors about delegated authority before the pipeline goes live.
-
The August Deadline Is a Documentation Deadline
High-risk obligations under the EU AI Act land in August. Almost nothing they ask for is a model property. Most of it asks who authorised what, and when.
-
Structured Output Isn't Reliable Output
JSON mode, function calling and constrained decoding give you schema compliance, not semantic reliability. Valid JSON can be completely wrong.
-
The Insurance Industry's AI Blind Spot: Claims Automation Without Trust Infrastructure
Insurance companies are racing to automate claims with AI. Nobody has built for the regulator, the litigant, or the appeals board. That is the blind spot.
-
What Moltbook Reveals About Multi-Agent Trust at Scale
Moltbook isn't an enterprise product - but the vulnerabilities it exposes matter for any organization deploying multi-agent AI systems.
-
The Agent Watchtower, Part 5: Reference Architecture
A complete, implementable design for enterprise agent governance. Concrete specifications, integration patterns, and implementation roadmap.
-
The Agent Watchtower, Part 4: Economics of Agent Operations
The financial model for sustainable AI governance. Cost cascading, ROI-driven routing, and why governance pays for itself.
-
The Agent Watchtower, Part 3: The Autonomy Spectrum
How to balance business unit freedom with enterprise governance. Federated control, trust-based permissions, and why guardrails beat gates.
-
The Agent Watchtower, Part 2: Anatomy of an Agent Control Plane
The technical architecture for unified agent governance: registry, observability, policy and control, and how they make multi-cloud governance possible.
-
The Agent Watchtower, Part 1: The Fragmentation Tax
Banks are deploying AI agents across AWS, Azure, GCP and open-source frameworks. The result: governance blind spots and a ticking regulatory problem.
-
AI Observability is Expensive Voyeurism
The observability market is selling you dashboards to watch your AI fail in high resolution. What you need is controllability.