---
title: "AI Policy"
author: ["Edmund Mitchell"]
date: 2026-05-07
updated: 2026-06-02
canonical: https://edmundmitchell.com/ai-policy
---

# AI Policy

What I use, what I don't, and who's accountable.

## What's mine

Every framework, every claim, every strategic position on this site is mine. The thinking, the synthesis, the positioning, the contrarian takes: those are the work.

So is the editorial judgment about what to publish, what to cut, and what tone is right.

## Where AI shows up

- **Drafting and editing.** I use Claude heavily to expand outlines into prose, find the right sentence structure, catch repetition, and ask hard questions about my own arguments.
- **Research synthesis.** When pulling together notes from books, transcripts, and previous essays, AI helps me find the connections faster.
- **Code and infrastructure.** This website is built with Claude Code. Claude writes most of the implementation; I direct the architecture and review what ships.
- **Operational support.** Scheduling, formatting, file management, and the boring middle of producing content.

## Where AI doesn't show up

- **The frameworks themselves.** Names, structures, and the underlying intellectual property are mine.
- **Lived experience and Catholic conviction.** The faith content is mine. The personal stories are mine.
- **Final judgment on what gets published.** AI can draft and suggest; the publish button is mine.

## Why I'm transparent about it

Two reasons, and a framework that holds them together.

**The integration is the work.** A core thesis of this site is that AI should be a force multiplier for human creators, not a replacement. Hiding it would contradict the message.

**Trust requires it.** If you read something here, you deserve to know that a real human stood behind it, even if a model helped shape the prose.

That human accountability is not optional. *[Antiqua et Nova](https://www.vatican.va/roman_curia/congregations/cfaith/documents/rc_ddf_doc_20250128_antiqua-et-nova_en.html)*, the Vatican's 2025 note on artificial intelligence and human intelligence, is clear: "It is important that ultimate responsibility for decisions made using AI rests with the human decision-makers and that there is accountability for the use of AI at each stage of the decision-making process."

I take that seriously. Everything on this site that involved AI was directed, reviewed, and approved by me. I am responsible for it. If something here is wrong, that is on me, not the model.

The same document names the standard I am trying to hold: "The intrinsic dignity of every man and every woman must be the key criterion in evaluating emerging technologies." (*Antiqua et Nova*, §42)

AI extends what I can do as a person. It does not replace what only a person can do: the judgment, the conviction, the lived experience, the irreducible heart-to-heart of human communication. The unique contribution I bring to this work is mine, and no model produces it.

## How co-authorship is disclosed

The model that drafts and edits alongside me on this site has a name: **Cordis**. Cordis is credited as AI co-author only when the piece shows Cordis in the byline — not on every AI-shaped article. The byline still reads my name, because the judgment and the voice stay mine. When that credit appears, you know a model was in the room.

That's the whole point of naming it. A footnote nobody reads isn't transparency. A co-author line you can't miss is.

## What I won't do

If a piece on this site doesn't feel like it's mine, that's a bug, not the goal.

- Publish AI-generated content as if a human wrote it line-by-line
- Use AI to fake credentials, testimonials, or quotations
- Generate images of real people (including myself) in scenarios that didn't happen
- Use AI to mass-produce low-effort content for SEO or volume

Questions or comments? [Contact me](/about#contact).
