The Journal
Essays on reading credentials before you pay for them.
The register gives you standings; the Journal gives you the reasoning. How to read a landing page, what a mark can and cannot carry, and what six hundred years of hallmarking quietly teaches about certifying people. Written by us, dated, and held to the same rule as everything else here: if we get something wrong, we say so, in the open, dated.
A Credential Is Not Compliance
The Journal · Reading time: 6 minutes Listen A credential is not compliance. Passing an exam proves a person studied a topic. It does not produce the record a regulator asks the organisation employing them to show. Article 96 of the EU AI Act does not ask a high-risk deployer to produce a certificate. It […]
Read it →The Relational Contract a Chatbot Can’t Sign
The Journal · Reading time: 7 minutes Listen The relational contract a chatbot can’t sign. Two real, documented cases, not a thought experiment: what happens when a system is built to feel like a relationship, and then behaves like a product. Everything else in this Journal is about a decision an organisation made and someone […]
Read it →Four Failures, Four Honest Disagreements
The Journal · Reading time: 8 minutes Listen Four failures, four honest disagreements. A B None of these four are stories about villains. Each is a real, documented system, a real harm, and a value someone chose over another value, in good faith, on a deadline. It is tempting to sort every AI failure into […]
Read it →Loophole Ethics
The Journal · Reading time: 7 minutes Listen Loophole ethics: why a code of conduct is not a person. THE GAP A written policy tells you what an organisation is willing to promise. It does not tell you what it does… the day that promise gets expensive. Past a certain size, every organisation writes one: […]
Read it →How to Read a “Certified AI Ethicist” Landing Page
The Journal · 3 Aug 2026 · six minutes How to read a “certified AI ethicist” landing page. The advertisement is real. The certificate is not, yet. The whole trick lives in the gap between those two sentences. Somewhere right now, a capable person is looking at a page with a shield on it. The […]
Read it →What Six Hundred Years of Hallmarking Can Teach AI Certification
The Journal · Reading time: 6 minutes Listen What six hundred years of hallmarking can teach AI certification. Six hundred years before anyone tried to certify an AI ethicist, London’s goldsmiths solved almost exactly this problem, for silver. In the 1300s, a buyer in London had no reliable way to tell a genuine silver spoon […]
Read it →What the Human Interior Actually Means
The Journal · Reading time: 6 minutes Listen What the human interior actually means. Every domain on the standard describes what you must know, do, or hold to. None of them describe what it costs to hold to it… Knowledge and skills are the parts of this job that show up on a syllabus: the […]
Read it →The Labour a Model’s Accuracy Score Never Mentions
The Journal · Reading time: 5 minutes Listen The labour a model’s accuracy score never mentions. Somewhere behind most large AI models sits a worker who spent a shift deciding whether a sentence or an image was too violent to leave in a training set. Her name is not on the model card. A model […]
Read it →The Nine Different Things “AI Regulation” Can Mean
The Journal · Reading time: 4 minutes Listen The nine different things “AI regulation” can mean. The EU AI Act and the NIST AI Risk Management Framework both get called “AI regulation” in casual conversation. A compliance team preparing for one would get almost nothing useful out of preparing for the other. UNESCO’s Consultation Paper […]
Read it →Ethics Washing, and How to Actually Spot It
The Journal · Reading time: 3 minutes Listen Ethics washing, and how to actually spot it. A set of AI principles costs an afternoon and a designer. A mandate that can actually stop a launch costs a budget line, a reporting relationship, and someone willing to push forward. The gap between those two is the […]
Read it →Explainable Is Not Interpretable
The Journal · Reading time: 3 minutes Listen Explainable is not interpretable, and the difference matters. “why?” A model can produce a fluent, specific-sounding reason for its own decision and still be a black box to the person asking. Those are two different claims, and vendors have every incentive to let you hear only one […]
Read it →An AI Ethics RFP That Isn’t Just a Checkbox
The Journal · Reading time: 3 minutes Listen An AI ethics RFP that isn’t just a checkbox. “Do you have an AI ethics policy?” is a question every vendor answers yes to, because it costs them nothing and proves nothing. A question that actually tests something is harder to write and much harder for a […]
Read it →When Your Role Spans Jurisdictions That Disagree
The Journal · Reading time: 3 minutes Listen When your role spans jurisdictions that disagree. A system built once and deployed everywhere does not get to obey one law. It has to satisfy the strictest applicable rule in every jurisdiction it touches, simultaneously, which is a harder engineering and governance problem than most organisations budget […]
Read it →AI Safety and AI Ethics Are Not the Same Field
The Journal · Reading time: Three minutes Listen AI safety and AI ethics are not the same field. Practitioner conversation treats “AI safety” and “AI ethics” as two labels for one field, interchangeable depending on which conference is hosting the panel. They are not the same field. They have different intellectual histories, draw from different […]
Read it →Does an AI Ethics Credential Actually Move a Career?
The Journal · Reading time: Three minutes Listen Does an AI ethics credential actually move a career? Dozens of AI ethics credentials now exist, issued by universities, professional bodies, and vendors of every size and seriousness, and almost none of the marketing for any of them answers the direct question a candidate actually wants answered: […]
Read it →Where do you actually stand?
Sixteen questions, six minutes. You rate yourself on each competence area, from "not yet" to "I could evidence this to an assessor". We hand back a profile, your strongest ground, and the gaps worth closing first. Nothing is stored unless you ask us to send it.