Specification

Decision procedure

Five questions, asked in order. The first one that terminates gives the level.

1. 1

   Was any generative AI used in making this work?

   Yes

   next

   No

   0

2. 2

   Did AI produce new material that appears in the published work — beyond mechanically processing what you had already made?

   Yes

   next

   No

   1

3. 3

   Did you make most of the final form yourself — most of the words, pixels, or sound?

   Yes

   2

   No

   next

4. 4

   Is the substance yours — the knowledge, the data, the argument, the direction?

   Yes

   3

   No

   next

5. 5

   Did a human review this before publication and take responsibility for what it says?

   Yes

   4

   No

   5

Surfaces

Optional. A single number describes the work as a whole. When parts of it differ sharply, declare them. The badge always shows one number; the breakdown lives behind the link.

Headline rule. The headline level describes the part of the work that carries its meaning. Decorative material that a reasonable audience would not consider material to the work does not drive the headline — declare it here instead.

Recognised surfaces: textimageaudiovideocodedata

{
  "aiUsageScale": "1.0",
  "level": 3,
  "surfaces": { "text": 3, "image": 5, "audio": 0 }
}

Edge cases

Every one of these came from trying to break the decision tree. They are settled here so that two honest people reach the same number.

Translation

A faithful translation inherits the level of the source work and adds a translation note. Translating a Level 1 article with a model publishes it as Level 1, translated by AI.

The substance and structure are unchanged. Penalising translation would make the world's writing less available, which is the opposite of the point.

Transcription

Transcribing speech you recorded is mechanical processing. It does not raise the level.

The words are yours. The model only wrote them down.

Substance from AI, wording from you

If a model originated the substance that materially determines what the work says — the thesis, the argument, the structure — it produced new material that survives into the work, even if you wrote out every word yourself. That is Level 2 or above, never Level 1. Feedback, critique, and fact-finding you verified stay Level 1: an editor's advice does not make the editor an author.

Ideas are material. Restating them in your own words does not un-invent them — and Level 1's promise that the model invented nothing has to mean it.

Decorative assets

A generated header image on a human-written article does not make the article Level 4. Declare the image under Surfaces.

The headline level describes what carries the meaning. Otherwise one stock image would swallow the whole declaration, and people would simply stop declaring.

Derivative works

Declare the level of your own contribution, and cite the level of the source if it declared one.

Live generation

Content generated on demand and shown to a user without review is Level 5, even if the system prompt was written carefully by a person.

Non-generative machine learning

Autofocus, denoise, upscaling, colour matching, and classification that invent nothing are not generative AI. They do not move you off Level 0.

If you doubt a declaration

A declaration is falsifiable in public. Check the link and the metadata; compare the declared level against the work and, where they differ, against the Surfaces breakdown; then ask the author — in a comment, a review, an issue. Most mismatches are honest readings of a boundary, and the edge cases above settle most of those.

If a declaration is knowingly false, the standard's answer is the one that scales: the claim was public and linkable, and now it is falsified in public. There is no detector and no registry here, on purpose — reputation is the enforcement, exactly as it is for a byline.

Interoperability

Mappings to other vocabularies are suggestions. They do not imply endorsement or support by Google, Schema.org, W3C, IPTC, or C2PA.

Level mapping

A level is a self-declaration. It does not certify accuracy, quality, rights, or legal compliance.

The Community Group's 9 October 2026 report uses human-only, ai-assisted and ai-autonomous for text. Levels 1 and 2 do not determine a value without the text's scope and actual review state. A Level 3 candidate requires the meaningful review in its full definition. Generic scale snippets omit the separate CG declaration so that publishers can assess it independently.

Machine-readable form

The canonical English definitions live in one file and drive the badges and machine-readable output. Localised explanatory copy is maintained separately and checked for structural parity.

Mappings to other vocabularies are suggestions. They do not imply endorsement or support by Google, Schema.org, W3C, IPTC, or C2PA.

GET /levels.json · CC0

<meta name="ai-usage" content="3">
<meta name="ai-usage-standard" content="https://usagescale.org">

<link rel="ai-disclosure" href="https://usagescale.org/3">

Keep the link to the public definition. Add optional metadata when your publishing system supports it.

{
  "aiUsageScale": "1.0",
  "level": 3,
  "surfaces": { "text": 3, "image": 0 },
  "definition": "https://usagescale.org/3"
}

Prior art

Nothing here is unprecedented, and pretending otherwise would be its own small dishonesty.