Home / Articles / Four frameworks, one agent / Versions / v1.1.0
Four frameworks, one agent: what changed in v1.1.0
From v1.0.0 (2026-10-13, e2ff68380) to v1.1.0 (2026-10-13, this release), paragraph by paragraph.
0 paragraphs added, 0 removed, 5 changed in place, 61 unchanged. About 12 words added and 1 removed. Insertions are marked like this, deletions like this; unchanged runs are folded to one line; figures appear as their file names.
23 unchanged paragraphs, under In short {#in-short}, The paper, and what it gets right {#paper}, The unit: a system before deployment, or one agent in one deployment {#unit}…
The OECD's five dimensions describe a system well, and three of them map straight across. People and Planet asks who the system touches and what it could cost them; that is the blast radius, which is defined over the reach because whoever takes an agent over inherits its reach and ignores its mandate. AI Model asks what the model is and how it learns; for the agents here the answer is short, the same model anyone else uses, which is why the risk sits in the deployment and not in the model. Economic Context asks how critical the system is and how far it spreads; the newsroom's honest answer, from The human in the loop is a spectrum, is that its side effects are small because its readership is small, and that this changes the position as readership grows.
4 unchanged paragraphs, under NIST: govern, map, measure, manage, as they run here {#nist}
Three rows deserve a sentence each. Govern 1, the policy, is only a policy if it survives the session: Re-anchoring found that a longthe session isbehind this site was summarised everyeighteen fewtimes daysin anda month, about 98% of its context replaced,replaced each time, so the rules live in a file and a hook prints them back after every summary. Measure 1 is the number I would put in front of any board: when the same customer service agent was built three ways, the share of its rules kept only by the model's good behaviour fell from 97% to 23%, and the policy shrank with it, because it no longer had to forbid what the agent could not do. And Manage 4 is where the paper's human oversight lives, which is the next principle.
8 unchanged paragraphs, under Japan: three roles, one policy between them {#japan}, The EU AI Act: the tier, the undefined word, and the dates {#act}
The second is a word. For a high-risk system, Article 9 of the Act requires the provider's risk management to leave residual risk "judged to be acceptable", and, as Every risk is already accepted points out, never defines acceptable. The paper restates the obligation without noticingremarking on the gap, which is fair,fair: becauseit almostis everyoneeasy does.to read past. Our answer is that accepted and acceptable are different things: one is a threshold, the other is an act by a named person on a date. A risk exists the moment the exposure does, so it is already being carried; the open questions are who has accepted it and until when. There are three doors, accept it for a stated interval, fund the work, or fix it, and silence is not a fourth: a risk nobody accepts rolls up to the holder's boss, and every path ends at the board. Every risk names the facts that make it true and the facts that would end it, down to the configuration file. That is what turns "judged to be acceptable" from a sentence into a record.
3 unchanged paragraphs, under The five principles on one small system {#principles}
So, the paper's five, each on the one system I can show, withand thehow honestfar status.each goes.
14 unchanged paragraphs, under What the graphs add, and the one to build next {#graphs}, Where it is hardest {#hardest}, What this does not claim {#limits}
• The system is small. One newsroom, one agent team, one person. The pattern is what I would defend, not the scale, and the paper is written forwith banks.financial institutions in mind.
9 unchanged paragraphs, under Where this comes from {#sources}