Asset Zero · What’s next
The framework is living
Version 1.0 froze on 31 August 2026 — and the framework’s own text says what happens next: it continues under AMC stewardship, section by section, and v1.x changes are made the way v1.0 was — proposed in writing, answered in writing, closed on rough consensus (§A1). This page is the worklist, the asks, and the open door.
The review cadence
The framework's review cadence — how often v1.x windows open — will be stated here by AMC and kept current. Until then: proposals are welcome any time through the discussion doors; they queue for the next review window.What the working group asked AMC to do — in its own order
Nine members ranked all nine Recommendations to Members in the final pass; this is the voted order, published top-first in §D2 (Borda rank sums over nine ballots — lower is better).
- 1.Adopt the Human–AI Boundary Frameworkrank sum 21six of nine first-place votes — the clear first
- 2.Publish the Decision Rights Matrixrank sum 36
- 3.Make "meaningful human oversight" verifiablerank sum 38tied on points with data quality; published third on the median rule — the tie is recorded in §D2
- 4.Treat data quality & provenance as a preconditionrank sum 38
- 5.Protect professional capabilityrank sum 39
- 6.Require an AI decision record and an AI-use registerrank sum 42
- 7.Set a "trigger gate" for any AI that actsrank sum 59
- 8.Adopt the 0–5 asset-management maturity scalerank sum 64
- 9.Govern vendor AI through procurement and assurancerank sum 68
Two asks recorded so they are not lost (§D2)
- Support for upskilling asset-management professionals in AI governance (the community's own words: “AI is here to stay — how can AM Council assist?”).
- A shared view of which AI tools and models members are actually using in the field today.
The tools the recommendations ask AMC to ship next
- A screening tool keyed to asset criticality (§C5)
- A procurement clause set for adaptation (§C4)
- A maturity self-assessment on the 0–5 path (§C5)
- The oversight-audit method — spot-check protocol with cadence and owner (OQ-02)
The v1.x worklist — carried openly
Twelve open questions ship inside the framework itself (the register, §D1), each with what would settle it. Four deserve naming here:
The eighth-principle question
The coverage question voted 6–3 to keep the candidate close-the-loop principle; the Project Chair withdrew his own candidate at the freeze, and the question returns to the working group in a v1.x cycle with the full record attached — where an adopted principle would carry a group vote as its provenance.
Affected parties
Do the people and communities an AI-influenced decision lands on need a principle of their own? 2–7 against in cycle 3; carried as a register field and challenge path, revisited after v1.0 — the majority option's own words (OQ-20).
Independent review
v1.0 makes no claim of independent review — the reviewer stream was parked after cycle 1, and the framework says so plainly (§A1). Standing the stream back up is a declared v1.x task.
Does the matrix move?
The baseline was set precisely so movement can be measured: re-running the matrix instrument on v1.x cycles and publishing the movement is OQ-22 — one of twelve open questions carried forward, each with what would settle it.
The lessons library grows from here
After v1.0 the failure-modes library (§C2) grows from members' own cases, contributed under the case-library opt-in: situation · where it broke · the fix · which principle would have caught it. Nine consented cases from cycle 1 seed it; cases are anonymised at the organisation level and never attributed without consent.
Events and endorsement
The launch event
Webinar 5 · The Boundary, Drawn — launching the framework
Launched Thursday 24 September 2026. The direction vote and volunteer register are open.
The vote and the register ↓Endorsement
v1.0 is signed by the Project Chair and published by the Asset Management Council; formal AMC Board endorsement is pending and will be recorded on the framework when given (§D4).
Where should this go next? Have your say
From the Project Chair
Version 1.0 was the floor, not the ceiling. One thing is decided: a plain-English companion is coming, and it will not change a ratified word. Six directions are on the ballot, and I want the vote to tell me where I am wrong. Independent eyes on v1.0 come before any of them. Tell me which of these matters to you, or name one I have missed. — Kai Dong, Project Chair
Iteration 2 · open to everyone
The direction vote and the volunteer register for iteration 2 are open to everyone until iteration 2 starts (early November, date to be confirmed). All you need is a free AMC account.
Already voted during the launch webinar? Your vote is recorded; no need to vote again. You can change it any time until iteration 2 starts.
The vote so far
Loading the tally…
The direction vote and the volunteer register for iteration 2 are open to everyone until iteration 2 starts (early November, date to be confirmed). All you need is a free AMC account.
Already voted during the launch webinar? Your vote is recorded; no need to vote again. You can change it any time until iteration 2 starts.
Comments, feedback, direction proposals — anything from “this section saved us a bad decision” to “v1.1 should tackle X”. Anyone signed in with an AMC account can post below; approved comments publish, and every input gets a written answer when its topic is next reviewed. Section-specific proposals belong on the framework page’s own doors.