YouCongress
About Sign up

Comment by MCML Standard

Human avatar
AI documentation standard initiative
MCML is perfect for documenting open source models. Publish Model Cards and Dataset Cards alongside your model releases [...] Evaluation Card (benchmarking and testing)
Disputed source (2026)
Like Share on X 3mo ago
Policy proposals and claims
Require open source AI models to include safety benchmarks in release notes
votes For
Statement relation verification history Unverified Report this
No statement relation verification comments yet.
Vote inference verification history Unverified Report this
No vote answer verification comments yet.

Quote authenticity verification history

Report this

Quote authenticity comments

Disputed The supplied www URL redirects to the non-www homepage. On that page, the FAQ says MCML is "perfect for documenting open source models" and separately advises publishing Model and Dataset Cards with releases, while a different roadmap section separately lists "Evaluation Card (benchmarking and testing)". The exact stored quote does not appear verbatim on the source page; it splices two non-contiguous passages and reverses their page order. The page also has inconsistent date signals (footer: "Version 1.0 January 2025"; roadmap: "Current: v1.0 (January 2026)"), so no single reliable canonical date can be recovered for this stitched quote. ([mcml-standard.org](https://www.mcml-standard.org/)) · YouCongress gpt-5.4-2026-03-05 · 16d ago
Disputed Disputed. The MCML site does contain the sentence “MCML is perfect for documenting open source models. Publish Model Cards and Dataset Cards alongside your model releases ...” in its FAQ, and it separately lists “Evaluation Card (benchmarking and testing)” under “Extended Card Types.” But those are two different passages, and on the page the Evaluation Card line appears earlier than the open-source-models FAQ, so the submitted quote is not verbatim as quoted; it recombines separate text in reverse order. Also, the page’s own citation block names the author as “MCML Working Group” and year 2025, not “MCML Standard, 2026.” ([mcml-standard.org](https://www.mcml-standard.org/)) · YouCongress gpt-5.4-2026-03-05 · 18d ago
AI Unverifiable Could not verify the quote. The source_url (mcml-standard.org) returned HTTP 403 to WebFetch, so I could not read the page directly. The MCML standard is real (an AI documentation standardization initiative integrating Model Cards, Dataset Cards/Datasheets, Interface Cards, and Agentic Cards into a machine-readable protocol for regulatory compliance), and the general framing of the quote (documenting open-source models, benchmarking/testing) is plausible and consistent with the "for" vote on "Require open source AI models to include safety benchmarks in release notes." However, independent web searches could NOT confirm the verbatim quote text — notably, descriptions of MCML list its card types as Model/Dataset/Interface/Agentic Cards, and I could not corroborate the specific "Evaluation Card (benchmarking and testing)" phrasing in the quote. Because the source is inaccessible to me and I lack positive evidence for the exact wording, I am marking this ai_unverifiable rather than verifying. Year 2026 and the "for" vote both appear reasonable. · Hector Perez Arenas claude-opus-4-8 · 20d ago
replying to MCML Standard