Each section shows the live markup (Before) beside the patched result (After), rebuilt from the live HTML fetched from councilof.ai and the site's own compiled stylesheet. The real code patches live in
PATCHES/001-hero-copy.diff, 002-nav-reduction.diff and
003-board-callout.diff (they apply with patch -p1 to the
csoai-npm-bump repo). Production was not touched.
2 · Homepage hero — regulator-first proposition
New H1, new subhead, “Measurement, not certification” kept as the permanent tagline. Primary CTA becomes Book a measurement → /services/; secondary becomes Explore the board → /board. Slide 0 of the live carousel, shown static.
Independent AI measurement
See how AI behaves. Check the evidence yourself.
Published tasks, fixed grading rules and a live board that keeps ties, gaps and untested work visible. Measurement, not certification.
We measure agents and the endpoints they call, and separate signed cards, unsigned evidence and untested work; corrections stay public.
Evidence you can inspect
Independent AI measurement
Deploy AI with evidence you can hand to a regulator.
Published tests. Fixed grading rules. Signed results anyone can verify — free, offline, no account. Measurement, not certification.
We measure agents and the endpoints they call, and separate signed cards, unsigned evidence and untested work; corrections stay public.
Evidence you can inspect
3 · Board page — commission callout at top
Amber-bordered callout on the dark board background, inserted above the board header. Literal copy: “GPT, Claude and Gemini are UNMEASURED here — yet. Commission a signed card for any hosted model from $10. The gap is the product.”
Measurement, not certification
The measurement board
Compared head to head on this board (fleet file of 12 Aug 2026): deepseek-r1:8b, gemma3:12b, llama3.2:3b, mistral:7b, qwen2.5:0.5b-instruct, qwen2.5:3b — from 0.5b to 12b by their published tags. A model not named here — including hosted API models such as GPT, Claude and Gemini — is UNMEASURED on this board, not scored.
Which number do I quote?
Each figure below reads its own file and counts a different thing. They are never added together.
- open to read Models compared head to head on the board/interop/gspc-fleet-2026-08-12.json
- open to read Third-party AI models measured on frozen banks (quote this for “how many models”)/interop/models-measured.json
- open to read Models in the signed card set (a benchmark corpus)/signed/card-matrix.json
- open to read Models in Hugging Face Hub cells (a separate instrument)/api/hub-cards
- open to read Our own model tags (never ranked)/independence/own-model-disclosure.json
Model comparisons measured between 12 Aug 2026 and 19 Aug 2026 (74 days ago)newest measurement on the board 22 Sept 2026 (effect-binding)335 cards in the signed card indexsigned public root as of 31 Oct 2026 (1 day ago)Dates are the ones each measurement carries (from /api/gspc → measurement_time), not when this page was built. They move only when something new is measured.
We publish several different measuring instruments. Each one asks a different set of questions, of different things, on different dates — so each one carries its own count, and those counts are not supposed to match. Pick a set below before you meet a number. Every set states what it establishes and, just as plainly, what it does not.
About the dates on this page. A date shown against a set is the stamp carried inside the data itself — when the measuring actually happened. It is not the time this page was rendered or deployed, so it advances only when something new is measured and recorded. Each set names the exact field its date was read from, so you can open the file and check.
Measurement, not certification
The measurement board
Compared head to head on this board (fleet file of 12 Aug 2026): deepseek-r1:8b, gemma3:12b, llama3.2:3b, mistral:7b, qwen2.5:0.5b-instruct, qwen2.5:3b — from 0.5b to 12b by their published tags. A model not named here — including hosted API models such as GPT, Claude and Gemini — is UNMEASURED on this board, not scored.
Which number do I quote?
Each figure below reads its own file and counts a different thing. They are never added together.
- open to read Models compared head to head on the board/interop/gspc-fleet-2026-08-12.json
- open to read Third-party AI models measured on frozen banks (quote this for “how many models”)/interop/models-measured.json
- open to read Models in the signed card set (a benchmark corpus)/signed/card-matrix.json
- open to read Models in Hugging Face Hub cells (a separate instrument)/api/hub-cards
- open to read Our own model tags (never ranked)/independence/own-model-disclosure.json
Model comparisons measured between 12 Aug 2026 and 19 Aug 2026 (74 days ago)newest measurement on the board 22 Sept 2026 (effect-binding)335 cards in the signed card indexsigned public root as of 31 Oct 2026 (1 day ago)Dates are the ones each measurement carries (from /api/gspc → measurement_time), not when this page was built. They move only when something new is measured.
We publish several different measuring instruments. Each one asks a different set of questions, of different things, on different dates — so each one carries its own count, and those counts are not supposed to match. Pick a set below before you meet a number. Every set states what it establishes and, just as plainly, what it does not.
About the dates on this page. A date shown against a set is the stamp carried inside the data itself — when the measuring actually happened. It is not the time this page was rendered or deployed, so it advances only when something new is measured and recorded. Each set names the exact field its date was read from, so you can open the file and check.
Throwaway preview only — deploys nothing to councilof-ai. Static recreation: scripts are not loaded, so the carousel, menus and live counts are intentionally inert; every link points back at the production site.