How I use AI in this studio
AI is a tool. A named one, with limits.
Large language models help me move faster on drafts, alt text, glossary cross-links, and spam triage. They don’t write the things you read here unattended, they don’t see anything you wouldn’t hand a freelancer with an NDA, and they don’t generate the photos in case studies. The ground rules are below, dated, and the changelog records every change to this page.
What AI does here
- First drafts of articles, glossary entries, and tool copy. The model produces a draft from an outline I write. I rewrite every paragraph for voice, fact-check every claim against the sourcing policy, add the inline citations, and date the result. The byline is mine because the words and the responsibility are.
- Alt text and image descriptions. The model proposes a description from the file; I trim and edit. Faster than typing 1,300 alt strings; same accessibility outcome.
- Search and cross-link suggestions. The model surfaces related glossary terms and articles for the “keep going” rails on tool result pages. The connection has to read true to me before it ships.
- Spam triage on inbound mail. Light classification only — flag obvious bot submissions for me to review. No autoreplies.
- Translation seeds. The Spanish mirror starts as a machine draft, then a fluent native-Spanish reviewer rewrites for register and idiom. Nothing ships under
/es/that hasn’t had a human pass. - Code, sometimes. Boilerplate, tests, and refactors. Every line is reviewed, run, and committed under my name. The generated code never includes data from a client engagement.
What AI never does here
- Train on your data. The free tools run in your browser; the inputs never reach a server, let alone a training corpus. And in Muntin Ledger — where you do hand over invoices — no language model reads them at all: extraction runs on Muntin’s own infrastructure, your invoice text is never sent to an outside AI provider, and a build gate blocks any release that would add a language-model library. The model that reads an invoice is fixed, so your corrections never train it. The commitment is the code, published at ledger.muntin.digital/promises with the enforcing scripts open to read.
- Sell prompts, embeddings, or transcripts. Nothing produced inside the studio — your invoices, your menu drafts, your meeting notes — gets resold, syndicated, or aggregated.
- Generate restaurant photos as real client work. The dishes in case studies are dishes a real photographer shot in a real kitchen. Hero shots are commissioned, not synthesized. Where AI imagery appears (illustrated diagrams, op-ed flat graphics), it’s labeled as illustration in the alt text.
- Write reviews, fake testimonials, or backdated “case studies.” Every review you read on this site has a person’s name and a verified context. Aggregate ratings only appear once five real ones exist.
- Make the call on a customer’s behalf. A model never sends a quote, signs off on a deliverable, or replies to your message as if it were me. The Window goes to my inbox; the answer is from me.
- Bypass the methods page. Any claim a draft surfaces gets a citation or it doesn’t ship. Hallucinated statistics are how a library loses operators — it’s the failure mode I most try to avoid.
Tools and providers
The current stack, on the record, with versions on the date below:
- Drafting and editing. Claude (Anthropic), GPT-class models (OpenAI). API mode with training opt-out enabled; consumer-mode chats are used only for non-confidential generic drafts.
- Code assistance. Claude Code. Same training-opt-out posture. No client repository contents are pasted into a consumer-grade UI.
- Embeddings. OpenAI embeddings power glossary cross-links and search ranking. Embeddings are computed on public site content only.
- What’s never here. No autonomous “agent” runs over client data. No model-generated emails sent without my review. No generative-fill on photographs of real food. And no language model anywhere in Muntin Ledger’s invoice-reading path — the providers above draft this website; they never see a customer’s invoice, and a CI gate keeps it that way.
Who reviews
Don Goldstein, the studio. Every page that shipped through an AI assist passed through my edit before publish. If I don’t agree with a sentence, it doesn’t make it to the page. The byline isn’t a model number.
How to cite this work
Quote it — just attribute and link. Name Muntin Digital and link the page you used; every figure in a library article is sourced inline in a cite drawer, so cite the same primary source I do (USDA, BLS, FRED, EIA, or the vendor’s own pricing page) where you can. A Cost Index number is a dated wholesale reference, not a delivered or retail price — say so, and link the ingredient page. The Cost Index data files at /cost-index/ are CC0; the writing is © Muntin Digital, quotable with attribution. There’s a machine-readable map for AI engines at /llms.txt, and every externally-verifiable number resolves to its primary source in /claims.json.
If you spot something off
A claim that smells synthetic, a citation that doesn’t resolve, a tone shift that reads like a model rather than a person — don@muntin.digital or The Window with topic=ai. Reply within four business hours; corrections show in the changelog within two business days.
Sister surfaces