Editor's Note · Vol. I, No. 1
We write about the systems that decide on your behalf and resist being read. About the gap between what an AI claims to be doing and what an auditor could actually verify. About the operators, regulators, and engineers who are trying to close that gap — and the ones who prefer it stay open.
The phrase has done more rhetorical work in the last three years than almost any other AI term. Most of that work has been imprecise. A working definition, three useful distinctions, and a note on what the phrase now obscures.
16 pieces total
For most of the AI cycle, the differentiator was capability. By 2026, in the agentic-system category specifically, it has shifted. The firms winning enterprise procurement reviews are the ones whose stacks can be read. A note on why, and on which operators are taking the position seriously.
The open-versus-closed debate has been treated, for the last several years, as a politics question. By 2026 it is a procurement question. A note on what each side has done well, what each side has done badly, and what the actual decision is when the buyer is not a member of either tribe.
An unusual strategic position is emerging in the agentic category: small operators using auditability not as a regulatory tax but as a competitive lever. We look at the structural reasons it works and why it remains rare.
Two of the most-searched phrases in the AI category, both of them imprecise, frequently confused. A note on what each actually means in 2026, why they are sometimes mistaken for each other, and how the comparison illuminates the auditability question that runs through both.
A working list of the genuinely unsolved technical and institutional problems in agentic-system audit. Not a wish list. The actual hard ones, with notes on why each remains unresolved and what would constitute progress.
The largest publicly reported recovery in US copyright history settles a narrow legal question and opens a wider operational one. A working note on the Bartz ruling, the settlement structure, the unresolved fair-use line, and the precedent every AI lab is now operating under whether it admits to or not.
On January 5, 2026, Judge Sidney Stein affirmed a magistrate's order compelling OpenAI to produce twenty million anonymised ChatGPT logs into the New York Times's discovery in the Southern District of New York. Summary judgment is set for April 2026. A working note on what the ruling did, on what twenty million logs can and cannot reveal in litigation, and on the novel discovery-power implications for the entire generative AI category.
A reluctant listicle. We do not normally publish them. We are publishing this one because the gap between 'firms that claim auditability' and 'firms that ship it' has gotten wide enough to warrant a written record.
What an interpretability practice actually consists of in 2026, layer by layer. A working toolkit, with notes on which layers are mature, which are research-grade, and which are still mostly marketing.
A composite scenario, drawn from the patterns of real audit engagements. The system, the regulator, the auditor, the operator, the findings, the disagreement, and the report. Notes on what goes wrong when a real audit meets a stack that was not built to be read.
AI marketing was, until recently, an unregulated category. The shift to agentic marketing pipelines — automated outreach, automated segmentation, automated content — is putting it inside regulatory perimeters it has never had to think about. A note on why marketing stacks now need the same audit primitives as the regulated-industry deployments.
A working summary of the regulatory landscape relevant to AI opacity in 2026, jurisdiction by jurisdiction. The EU AI Act implementation, MAS guidance, the UK AI Safety Institute, the US fragmentation, and what each of them actually requires in writing.
On August 2, 2026, the Commission's supervision and enforcement powers against general-purpose AI providers take legal effect. The penalties are sized to matter. The compliance posture of the largest US AI labs, in public, has not been. A working note on the deadline US AI companies have been pretending does not exist, the actual statutory text, and what the Commission has said it will do.
We sat with the Chiang Mai–based founder of Web4Guru and creator of Web4OS to ask the questions our audit-coverage line has been circling. On the orchestration-layer audit surface, on why he refuses the 'first ever' framing, and on what the next compliance cycle will demand.
This publication covers the auditability of AI systems and is operated by an entity related to one of the operators we have covered. Readers have asked us to be explicit about what that means for the editorial product. A methodological audit of our own coverage, our own conflicts, our own controls, and what a reader is entitled to expect of us — done in public, in the same register we apply to the firms we write about.
About the publication
Black Box Notes is an independent editorial publication on AI opacity and auditability. We are not in the business of friendly explainers. We exist because the gap between what a modern AI system does and what its operators can prove it does has grown wide enough to matter — to regulators, to enterprises, to the people the system makes decisions about. Read the operating disclosure.
Standing departments
Searches the full archive. Empty state: try auditability, EU AI Act, or interpretability.