Perspectives on AI-native operating models, agentic work, decision intelligence, adoption and responsible scale.
PUBLISHED
The latest we’ve written.
Does your board understand the AI it approves?
A board recently approved a fraud detection model with the same ceremony it uses for an acquisition: committee, slide deck, the usual due diligence questions. Nobody asked how rare the fraud the model was built to catch actually is. Without that number, the accuracy figure on the slide meant nothing.
26 September 2026 · 7 min
Why Teams Revert to Manual Work After AI
You bought the tool, rolled it out well, and it worked. For months. Until one agent got something wrong exactly once, at the worst moment, and ever since half the team does the task twice: with the agent, to comply, and by hand, just in case.
25 September 2026 · 8 min
How Many AI Agents Can One Person Run?
The number every staffing committee reaches for is fixed: four, twelve, twenty. The one that actually matters depends on a variable nobody puts on the slide.
24 September 2026 · 7 min
Who's liable when your AI vendor gets breached
The AI vendor that screens the company's job applicants sends a note on a Friday evening: unauthorized access to a database. The data processing agreement has been signed for two years. Nobody yet knows if that piece of paper covers what happens next.
19 September 2026 · 6 min
Who Owns a Process You Run With an Agent?
An insurer paid a claim it should never have paid. The agent had pre-approved the case with the full file attached; the person who signed the final approval checked the amount, not the cause. When risk asked who had decided to pay that claim, IT said the agent only recommends, operations said the person signs whatever lands on their desk, and nobody else wanted to answer.
18 September 2026 · 6 min
The missing AI-native hire isn't technical
Every job posting says the same thing: young engineer, AI-fluent, practically born using a prompt. Hire ten of them and the company runs exactly the way it did before, because that was never the role that was missing.
17 September 2026 · 6 min
The KPI That Never Changed When the Agent Arrived
A support team hit its quarterly target for the first time all year: cases resolved per person, up. Nobody asked how many of those cases the agent drafted and how many the person actually solved. The bonus went out anyway.
16 September 2026 · 6 min
How to Avoid Getting Locked Into an AI Vendor
The vendor changed the terms at the year-two renewal. Procurement asked for competing quotes and found something worse than the new price: switching would take longer than replacing the accounting system. They signed.
15 September 2026 · 6 min
What to Do About the AI Your Team Already Uses
A company's security team cross-referenced its own network logs and found that a good part of the staff was already uploading internal documents to an AI nobody had approved. The response was to block access that same afternoon. Two weeks later the traffic came back. Nobody could see it anymore.
14 September 2026 · 7 min
What happens to managers when agents do the work
An operations manager kept the same team, the same title, and half the caseload moved to an agent. Nobody shrank the team, nobody changed the title, and nobody sat down with him to decide what happens when the agent gets a case wrong that he never saw.
13 September 2026 · 7 min
How to Justify an AI Investment With No ROI Yet
The playbook for justifying AI's return is already written, and it gets repeated in every proposal that reaches a board. The problem isn't the metric. It's the director who already watched two pilots die and won't be talked into another slide of indicators.
12 September 2026 · 7 min
When AI Layoffs Force You to Rehire
Two in three companies that laid off staff because of AI are already rehiring some of those roles. The question nobody asked before signing the layoffs wasn't whether the AI worked. It was exactly which task it would handle and which one it would not.
11 September 2026 · 6 min
What Really Happens If You Fall Behind on AI
Fear of falling behind pushes companies to buy anything just to say they started. That move sets a company back further than its real distance from competitors.
10 September 2026 · 6 min
Why a Multi-Agent System Fails in Production
Every agent passed its test on its own. Together, in production, the process started failing, and when someone asked which of the three had broken, the answer was none of them.
9 September 2026 · 6 min
Who Owns What an AI Agent Creates for a Company?
A law firm sent a client a proposal drafted mostly by an internal agent trained on the firm's own past files. Before signing it, a partner asked who owned the document. Nobody in the room had an answer. The firm had run the tool for two years and never once asked out loud.
8 September 2026 · 7 min
How to Choose Between AI Vendors That Look Alike
AI vendors don't compete on being different. They compete on sounding identical, and that tie is deliberate.
7 September 2026 · 6 min
What an AI Agent Really Costs After You Sign
The licence is the number the committee debates. The one that actually matters, what it takes to sustain the agent once it's live, never makes it into a proposal.
6 September 2026 · 6 min
Leading AI Isn't a Title, It's a Stage
A management committee at a two-hundred-person company read the same piece everyone's been sharing on LinkedIn: a comparison of Chief AI Officer, Chief Data Officer and Chief Information Officer, explaining which one fits which size of company. None of the three fit this year's headcount budget.
5 September 2026 · 7 min
Supervise an Agent Without Redoing Its Work
"Just supervise it" is the instruction that ends with someone opening every case by hand. Nobody defined what supervising means, so everyone picks the option that leaves them covered: check everything.
4 September 2026 · 6 min
What AI-Freed Time Actually Costs You
An operations director told me his team gained half a free day a week once the weekly reports ran themselves. I asked what they had done with that half day. He went quiet. Nobody had decided. It filled itself in, with something else.
3 September 2026 · 6 min
How to Audit a Decision an AI Model Made
An auditor asked why an AI model, bought from a vendor, made a particular call. What came back was a confidence score. A percentage is not a reason. It is a measure of how sure the model was that it could be wrong just the same.
2 September 2026 · 7 min
Build, Buy, or Partner: How to Decide
Almost no AI buying committee asks the right question. It asks the price. It should ask how long the edge it's signing for will last.
1 September 2026 · 7 min
Where a well-built AI strategy dies
The strategy got approved without a single dissenting vote. It had a value-and-feasibility matrix, a prioritised portfolio of initiatives and a full chapter on data governance. Six months later there was no trace of it in any document still in use, and nobody on the board could say why.
31 August 2026 · 7 min
How the Operating Model Changes With Agents
They redrew the org chart. They redesigned the billing process task by task. And they still convene the same exception committee every Tuesday at ten, even though the agent watching that queue already closed three hundred decisions since Monday.
30 August 2026 · 7 min
What makes a company AI-native, not its product
Almost every company I meet has AI somewhere in the product. Very few have changed how they decide anything. That gap is the one that matters, and you cannot see it by looking at the product.
29 August 2026 · 6 min
Where a board should start with AI
The first move for almost any board that decides to take AI seriously is naming a committee. It is also the move that resolves the least: it relocates the question instead of answering it.
28 August 2026 · 6 min
Governing AI without slowing the business
The AI governance committee has spent six weeks reviewing a case any analyst would clear in two days. Nobody calls that slowing the business down. It is exactly that, and nobody has measured it.
27 August 2026 · 6 min
What Changes in Structure With Real AI
Every AI restructuring plan repeats the same line: control moves from IT to the business. It sounds like a decision. In practice it fits any slide, because it never forces anyone to move a single box on the org chart.
26 August 2026 · 6 min
What controls does a board need before scaling AI
The risk committee asked for an AI data policy and had it in three days. What it still doesn't have, six months later, is an answer to the question that actually matters: what happens when the control says no and the business decides to proceed anyway.
25 August 2026 · 6 min
How to measure the real ROI of AI
In the boardroom someone puts up a slide that says we saved time, and nobody asks the question that matters: compared to what. Without that answer, the number measures nothing.
24 August 2026 · 6 min
What Can an Agent Do Without Oversight
The "classify by risk" framework sounds like a method. In practice nobody in the room can apply it to a real case until it has already gone wrong.
23 August 2026 · 7 min
Why your team won't use the AI you gave them
You bought the licences. Three months in, the usage panel says the thing nobody puts on the slide.
22 August 2026 · 6 min
How to redesign a process for an AI to run
They dropped an agent into the middle of the process and kept doing exactly the same thing, just with one extra step: someone copies what the agent says into the usual spreadsheet, just in case. Redesign is not that: it is running every task in the process through a five-question filter before deciding who runs it.
21 August 2026 · 6 min
Why your AI pilots never reach production
The pilot worked. And that is precisely the moment it stopped moving.
20 August 2026 · 7 min
Who answers when an agent answers
The question that decides whether an agent reaches production isn’t technical. It’s who owns the outcome when it goes wrong.
20 August 2026 · 5 min
EDITORIAL PILLARS
What we’re going to write about, and what we won’t.
Five lines. No model news, no tool comparisons, no summaries of what someone else already said.
The AI-native enterprise
Strategy, ambition and enterprise design. What sets an AI-native company apart from a company that just uses AI.
Agentic work
Workflows, roles, agents and human accountability. How work changes when part of it is done by a system.
Operating-model reinvention
People, Data, Agents, Products and Operations. The design decisions that determine where value accumulates.
Value and adoption
Measurement, trust, change and capability transfer. Why adoption is a design problem, not a communication one.
Responsible scale
Governance, controls, risk and scale readiness. When to scale and, above all, when not to.