---
title: "Override doctrine | Len P. van der Hof | Len P. van der Hof"
description: "A charter says what an agent does. An override doctrine is the written list of what it must refuse: spend, speak as the company, change production, or close a judgment you still own."
image: "https://lenvanderhof.com/media/generated/blog-hero-override-lockout-night-v1.44254d337fb9.wide.webp"
---

[AI Systems](https://lenvanderhof.com/en/blog/category/ai-systems/) Research guide

# What an agent may never do

HITL is who can fail the output. The doctrine is what must be refused when no one is watching.

Len P. van der HofPublished 23 August 20263 min read

The wheel does not turn. That is the doctrine.

Direct answer

Override doctrine is the written list of decisions an agent may never take. The live glossary names four: spending above a ceiling, speaking as the company, changing production data, and closing a judgment the founder still owns. It is the complement of a charter. The charter says what the agent does. The doctrine says what it must refuse. Write it before the agent gets a tool that can send, spend, or delete. A named human in the loop is not a substitute. If the irreversible act already happened, the loop is decorative. CHORUS Review is the gate that enforces the list. The list is not the six-skill protocol.

## Key takeaways

- The charter is not the doctrine. If you only wrote what the agent does, you wrote half.
- Four refusals, already public: spend above a ceiling, speak as the company, change production data, close a judgment you still own.
- Write it before the tool that can send, spend, or delete. Folklore in Slack is not a doctrine.
- HITL without a refuse-list arrives after the spend. The doctrine has to hold at 03:00.

The team wrote a charter. Three pages of what the agent *does*. Nothing says what it must refuse. At 03:00 it sends the customer the draft, because the draft was ready and no line told it to stop. The usual diagnosis is “we need better HITL.” Often the human was never given a written refuse-list.

## The charter is not the doctrine

A [charter](https://lenvanderhof.com/en/blog/what-is-an-ai-agent/) says what the worker does. Purpose, inputs, outputs, a ceiling, a review. That page already exists.

[Override doctrine](https://lenvanderhof.com/glossary/override-doctrine/) is the other half. It is the written list of decisions an agent may never take. If you only wrote the does-list, you wrote half. The half that fails at 03:00 is the missing half.

This is not a personality test for the model. Models do not have temperament. They have tools. The doctrine is about the tools.

## Four refusals, already public

The live capsule names four. I am not adding a fifth to make a framework.

**Spending above a ceiling.** Draft the invoice. Do not send the money.

**Speaking as the company.** Draft the reply. Do not put the company’s name on a channel the human has not opened.

**Changing production data.** Propose the patch. Do not write production.

**Closing a judgment you still own.** Assemble the brief. Do not decide the hire, the price, or the kill.

Those four are irreversible or identity-bearing. An agent can work up to the line. A human crosses it. [The Multi-Agent Organization](https://lenvanderhof.com/books/multi-agent-organization/) puts the same cut in the charter field it calls the authority ceiling: keep the agent on the reversible side. Default to deny. That book is live. This page is the list, not the seven-field template.

## Write it before the tool

The glossary is specific about *when*. Write the doctrine before the agent gets a tool that can send, spend, or delete. [MCP](https://lenvanderhof.com/en/blog/mcp-explained-for-founders/) is how a step touches the building. The doctrine is who may not turn the key.

If the refuse-list lives in a Slack thread from March, it is not a doctrine. It is folklore. Folklore does not bind a tool call.

## HITL is not a substitute

A [human in the loop](https://lenvanderhof.com/glossary/human-in-the-loop/) is a named person with a ceiling and a gate. A glance at a dashboard is not a loop.

HITL without a doctrine arrives after the spend. The human is then downstream of a decision that already happened. That is not a loop. That is an autopsy.

The doctrine has to hold when no one is watching. If the only control is a person who sleeps, you did not write a refuse-list. You wrote a hope.

## Review is the gate, not the list

[CHORUS](https://lenvanderhof.com/frameworks/chorus/) is the coordination protocol for teams of two to fifteen. Review is one of six skills. It is how the team fails an output against written criteria.

This page is not a second CHORUS article. The protocol is the gate. The doctrine is the criteria the gate reads. If Review has nothing written to fail against, Review is a vibe.

## A fifteen-minute audit

Pick the agent that can send, spend, or delete.

1. Paste the written refuse-list. If you have to remember it, it is not written.
2. Check the four public refusals. Circle the ones that are missing.
3. Name the last irreversible action this agent took. Whose name did it carry?
4. Write the file the doctrine lives in. Not the chat.
5. Write the action you will take off the agent today if step 1 is blank.

If step 1 or 4 is empty, take the tool away. A charter without a doctrine is a permission slip. Write the refusals. Then the 03:00 call has something to hit.

## Terminology

- [Override doctrine](https://lenvanderhof.com/glossary/override-doctrine/)
- [Human in the loop](https://lenvanderhof.com/glossary/human-in-the-loop/)
- [CHORUS](https://lenvanderhof.com/glossary/chorus/)

## Sources

1. [Override doctrine (glossary)](https://lenvanderhof.com/glossary/override-doctrine/)
2. [Human in the loop (glossary)](https://lenvanderhof.com/glossary/human-in-the-loop/)
3. [CHORUS framework](https://lenvanderhof.com/frameworks/chorus/)
4. [What is an AI agent?](https://lenvanderhof.com/en/blog/what-is-an-ai-agent/)
5. [MCP explained for founders](https://lenvanderhof.com/en/blog/mcp-explained-for-founders/)
6. [The Multi-Agent Organization](https://lenvanderhof.com/books/multi-agent-organization/)

## Further reading

- [Override doctrine](https://lenvanderhof.com/glossary/override-doctrine/)
- [Human in the loop](https://lenvanderhof.com/glossary/human-in-the-loop/)
- [CHORUS](https://lenvanderhof.com/frameworks/chorus/)
- [What is an AI agent?](https://lenvanderhof.com/en/blog/what-is-an-ai-agent/)
- [MCP explained for founders](https://lenvanderhof.com/en/blog/mcp-explained-for-founders/)
- [The Multi-Agent Organization](https://lenvanderhof.com/books/multi-agent-organization/)

About the author

## [Len P. van der Hof](https://lenvanderhof.com/en/authors/len-p-van-der-hof/)

Entrepreneur, AI Innovator and Venture Builder

Len P. van der Hof builds practical AI systems, digital ventures and evidence-informed tools for founders.

```json
{
	"@context": "https://schema.org",
	"@graph": [
		{
			"@type": "Person",
			"@id": "https://lenvanderhof.com/#person",
			"name": "Len P. van der Hof",
			"alternateName": [
				"Len van der Hof",
				"L.P. van der Hof",
				"Leendert Pieter van der Hof"
			],
			"honorificSuffix": "MSc",
			"url": "https://lenvanderhof.com/",
			"image": [
				"https://lenvanderhof.com/photos/len-portrait-1.jpg",
				"https://lenvanderhof.com/photos/len-portrait-2.jpg",
				"https://lenvanderhof.com/photos/len-portrait-3.jpg",
				"https://lenvanderhof.com/photos/len-portrait-4.jpg",
				"https://lenvanderhof.com/photos/len-speaking.jpg",
				"https://lenvanderhof.com/photos/len-hero.jpg"
			],
			"jobTitle": "Entrepreneur, AI Innovator and Venture Builder",
			"description": "Len P. van der Hof is a Dutch entrepreneur and AI innovator building ReasonKit, MindSesh and technology ventures through LPH98.ventures.",
			"address": {
				"@type": "PostalAddress",
				"addressLocality": "Zwijndrecht",
				"addressCountry": "NL"
			},
			"alumniOf": {
				"@type": "CollegeOrUniversity",
				"name": "Rotterdam School of Management, Erasmus University"
			},
			"knowsAbout": [
				"Artificial intelligence",
				"AI agents",
				"Agentic AI systems",
				"SEO",
				"Generative engine optimization",
				"Answer engine optimization",
				"Venture building",
				"Founder performance",
				"Founder psychology",
				"Evidence-based decision-making"
			],
			"sameAs": [
				"https://www.linkedin.com/in/lenvanderhof/",
				"https://x.com/LenvanderHof",
				"https://www.youtube.com/channel/UCTG20buKqYYbitqqf7l3zJA",
				"https://www.instagram.com/Lenvanderhof/",
				"https://www.threads.com/@lenvanderhof",
				"https://github.com/Lenvanderhof",
				"https://www.goodreads.com/author/show/70983905.Len_P_van_der_Hof",
				"https://www.amazon.com/stores/author/B0H8B1SCLQ",
				"https://www.bol.com/nl/nl/b/len-p-van-der-hof-msc/609879394/",
				"https://bsky.app/profile/lenvanderhof.com",
				"https://mastodon.social/@Lenvanderhof",
				"https://crates.io/users/Lenvanderhof",
				"https://cursor.com/@Lenvanderhof",
				"https://medium.com/@Lenvanderhof",
				"https://gitlab.com/Lenvanderhof",
				"https://hub.docker.com/u/lenvanderhof/",
				"https://dev.to/lenvanderhof",
				"https://www.facebook.com/Lenvanderhof",
				"https://soundcloud.com/Lenvanderhof"
			],
			"affiliation": [
				{
					"@id": "https://lenvanderhof.com/#publisher"
				},
				{
					"@id": "https://lenvanderhof.com/#mindsesh"
				}
			]
		},
		{
			"@type": "WebSite",
			"@id": "https://lenvanderhof.com/#website",
			"url": "https://lenvanderhof.com/",
			"name": "Len P. van der Hof",
			"description": "Len P. van der Hof is a Dutch entrepreneur and AI innovator building ReasonKit, MindSesh and technology ventures through LPH98.ventures.",
			"inLanguage": [
				"en",
				"nl"
			],
			"publisher": {
				"@id": "https://lenvanderhof.com/#person"
			}
		},
		{
			"@type": "Organization",
			"@id": "https://lenvanderhof.com/#publisher",
			"name": "LPH98.ventures",
			"url": "https://lph98.ventures",
			"founder": {
				"@id": "https://lenvanderhof.com/#person"
			}
		},
		{
			"@type": "Organization",
			"@id": "https://lenvanderhof.com/#mindsesh",
			"name": "MindSesh",
			"url": "https://mindsesh.net",
			"founder": {
				"@id": "https://lenvanderhof.com/#person"
			}
		},
		{
			"@type": "SoftwareApplication",
			"@id": "https://lenvanderhof.com/#reasonkit",
			"name": "ReasonKit",
			"url": "https://reasonkit.sh",
			"creator": {
				"@id": "https://lenvanderhof.com/#person"
			}
		},
		{
			"@type": "ImageObject",
			"@id": "https://lenvanderhof.com/en/blog/override-doctrine/#primaryimage",
			"url": "https://lenvanderhof.com/media/generated/blog-hero-override-lockout-night-v1.44254d337fb9.wide.webp",
			"contentUrl": "https://lenvanderhof.com/media/generated/blog-hero-override-lockout-night-v1.44254d337fb9.wide.webp",
			"representativeOfPage": true
		},
		{
			"@type": "BreadcrumbList",
			"@id": "https://lenvanderhof.com/en/blog/override-doctrine/#breadcrumb",
			"itemListElement": [
				{
					"@type": "ListItem",
					"position": 1,
					"name": "Home",
					"item": "https://lenvanderhof.com/"
				},
				{
					"@type": "ListItem",
					"position": 2,
					"name": "Blog",
					"item": "https://lenvanderhof.com/en/blog/"
				},
				{
					"@type": "ListItem",
					"position": 3,
					"name": "AI Systems",
					"item": "https://lenvanderhof.com/en/blog/category/ai-systems/"
				},
				{
					"@type": "ListItem",
					"position": 4,
					"name": "What an agent may never do",
					"item": "https://lenvanderhof.com/en/blog/override-doctrine/"
				}
			]
		},
		{
			"@type": "WebPage",
			"@id": "https://lenvanderhof.com/en/blog/override-doctrine/#webpage",
			"url": "https://lenvanderhof.com/en/blog/override-doctrine/",
			"name": "What an agent may never do",
			"description": "A charter says what an agent does. An override doctrine is the written list of what it must refuse: spend, speak as the company, change production, or close a judgment you still own.",
			"isPartOf": {
				"@id": "https://lenvanderhof.com/#website"
			},
			"primaryImageOfPage": {
				"@id": "https://lenvanderhof.com/en/blog/override-doctrine/#primaryimage"
			},
			"breadcrumb": {
				"@id": "https://lenvanderhof.com/en/blog/override-doctrine/#breadcrumb"
			},
			"inLanguage": "en-GB"
		},
		{
			"@type": "BlogPosting",
			"@id": "https://lenvanderhof.com/en/blog/override-doctrine/#article",
			"mainEntityOfPage": {
				"@id": "https://lenvanderhof.com/en/blog/override-doctrine/#webpage"
			},
			"headline": "What an agent may never do",
			"description": "A charter says what an agent does. An override doctrine is the written list of what it must refuse: spend, speak as the company, change production, or close a judgment you still own.",
			"datePublished": "2026-08-23T07:00:00.000Z",
			"author": {
				"@id": "https://lenvanderhof.com/#person"
			},
			"publisher": {
				"@id": "https://lenvanderhof.com/#person"
			},
			"image": [
				"https://lenvanderhof.com/media/generated/blog-hero-override-lockout-night-v1.44254d337fb9.square.webp",
				"https://lenvanderhof.com/media/generated/blog-hero-override-lockout-night-v1.44254d337fb9.landscape.webp",
				"https://lenvanderhof.com/media/generated/blog-hero-override-lockout-night-v1.44254d337fb9.wide.webp"
			],
			"articleSection": "AI Systems",
			"inLanguage": "en-GB"
		}
	]
}
```
