Release 1.3, standards submissions and independent pilots #11

Release and public submission record for the October 2026 upgrade. The maintainer requested publication and public submissions on 11 October, followed by corrections to the reference-page layouts.

Published releases

• v1.3.0: typographic redesign, corrected content and metadata, README, schemas, packaging and security tooling. Merged through PR #12.
• v1.3.1: seven consistent document views, aligned reading columns, desktop/mobile contents, readable mappings and code, shorter project status and RTL corrections. Merged through PR #24.
• CI, CodeQL, deployment and release packaging passed for commit 37bc433395dc2e9369f852aa250af8f2874e2cba.
• Production at https://usagescale.org passed 40 HTTP/data checks, including 22 language vocabularies, both schemas, document pages, CSP and retired signing redirects. Chrome checked nine document routes at desktop and mobile widths, keyboard contents controls, clipboard and code direction; no browser/CSP errors were recorded.
• Downloaded release archives match GitHub's SHA-256 digests. The vocabulary ZIP includes both schemas and 54 marks. The MCP archive includes its classifier and licence; an isolated installation of the downloaded archive passed the protocol tests.

The release archives are ai-usage-scale-v1.3.1.zip and usagescale-mcp-1.3.1.tgz. The MCP package is distributed from source and release archives; it has not been published to npm.

Repository configuration

• Updated the description and structured-data topics; clarified the MIT/CC0/CC BY licence scopes in the README and licence documents. GitHub now recognises MIT for the code.
• Enabled Dependabot vulnerability alerts and automatic security updates. CodeQL is active. No open Dependabot or main-branch CodeQL alerts were recorded in the final check.
• Preserved protected main and version tags; releases went through the required checks.
• Closed superseded Dependabot PRs Bump fast-uri from 3.1.3 to 3.1.8 #13–Bump sharp and astro #17 after confirming their target versions were already present on main. The remaining dependency-update PRs are separate maintenance proposals.

Public submissions

• Schema.org #3391: contribution and review use cases.
• W3C AI Content Disclosure CG #51: related-work request.
• First Schema.org feedback from robrwo, 11 October and our follow-up: some projects do not need distinctions in the amount of AI use, and that amount is difficult to establish objectively. We acknowledged that this concern applies to the scale's Level 2 majority threshold and asked whether optional human-review information would be useful.
• Record further responses, resulting changes and any acceptance separately.
• Record independent production pilots with public evidence.

These submissions request consideration; they do not establish standards acceptance or platform support. The W3C request concerns related work. Substantive specification contributions follow the group's membership and contribution process.

Independent pilots and review

Publishers and tool maintainers can propose a pilot here with the type of work, declared level, original human contribution, review process and a public declaration URL. Mixed-media workflows and the distinction between original human substance and generated form are useful comparison cases. There is no registration requirement or fee.

The existing CHAOSS references and deferred diarie integration interest are recorded in ADOPTERS.md. The October audit did not verify an independent production declaration.

The first Schema.org response also raises the need to validate the Level 2 majority boundary across real publishing workflows. Pilots should examine whether different publishers reach consistent classifications from the same process records. The scale remains a self-declared process vocabulary; a level does not independently verify the account. A basic AI-use disclosure may be sufficient for some use cases, with richer contribution/review information optional.

The Level 3 definition requires meaningful review, but its current decision path stops before asking about review. Automated Level 3 results remain provisional. Human editorial review of the revised website and fluent review of the 21 machine translations remain pending; publication does not change that status.

Google

Google's Search feature gallery does not document a dedicated feature for this vocabulary. The actual submissions above use the verified Schema.org and W3C public venues.

Google also links an SEO Office Hours question form from its official instructions. This accepts general questions about Google Search. The form currently requires a mathematical verification; that step must be completed manually. No submission through this form has been sent.

• Submit the prepared question manually through the official form and record confirmation.

Your question:

What is the appropriate public route for proposing standardized disclosure of generative AI contribution and meaningful human review for Google Search? What implementation evidence would be useful before Google considers support for a disclosure vocabulary?

Additional context:

I maintain AI Usage Scale, an independent CC0 vocabulary with six levels: https://usagescale.org. We have submitted contribution/review use cases to Schema.org #3391 and a related-work request to the W3C AI Content Disclosure Community Group #51. These are requests for consideration; we do not claim standards acceptance or Google Search support. We would welcome guidance on the public process and evidence that publishing-tool authors should provide.

The optional name field can identify Konrad Karauda. No Twitter username is needed. Any answer or resulting product support will be recorded here separately.