#board-governance

How Boards direct, oversee, and stay accountable as AI lands.

Tagged articles (75)

A course inspector in a waxed jacket stands alone on an empty, mist-covered steeplechase course at dawn, one hand resting on the top of a birch fence she is checking before any horse runs, an electric blue running rail curving away behind her into the mist: the checks an agent must pass are made by people before the run, not during it. (Image generated by ChatGPT 5) — Humans on the Loop: Governing What an Agent Does, Before It Does It
8 November 2026 · 11 minute read
Agents act inside rules written before they run. How much discretion an agent is given, and on what terms, is what the Board approves.

A lock keeper stands at a closed canal lock gate in golden evening light, his hand on its electric blue winding gear while a narrowboat waits in the chamber below, embodying the argument that an agent's model can only propose a journey and that whether it proceeds is decided by a check outside the model, enforced at the point where a proposal becomes an action. (Image generated by ChatGPT 5) — The Model Did Not Phone Home: Where an Agent's Actions Are Decided
1 November 2026 · 12 minute read
Cohere's test shows an agent's actions are decided by the harness around the model. Governance belongs where a proposal becomes an action.

A craftswoman in a sunlit morning workshop lifts one small screwdriver with an electric blue handle from a wall of tools graded from delicate to heavy, leaving the larger tools on their hooks, embodying the Principle of Least Intelligence: the mature choice is the least capability that does the job, not the most powerful tool available. (Image generated by ChatGPT 5) — The Principle of Least Intelligence: How Much AI Does the Work Need?
11 October 2026 · 14 minute read
Apply no more intelligence than the outcome needs: the Principle of Least Intelligence matches capability to the work, with permission to act decided separately.

A manager drawing an organisation chart in electric blue marker on the glass wall of a daylit meeting room, a neat row of boxes for the team sitting above a second, longer row of lines running to boxes that no longer fit on the glass, embodying the argument that the manager's span now includes agents nobody counted, and that the AI business case becomes real or fails in that one job. (Image generated by ChatGPT 5) — Managers Own the Workflow: Where the AI Business Case Becomes Real
4 October 2026 · 10 minute read
Most AI business cases count the time agents save; the saving becomes real, or not, in the manager's workflow, a job rarely redesigned for it.

A senior director stands in a daylit corridor looking through a clear glass wall into a boardroom where a group sits around a table beneath an electric blue display, nothing hidden and nothing he can join, embodying the argument that decisions about who gets frontier capability are taken openly by suppliers and governments in rooms the customer is not party to. (Image generated by ChatGPT 5) — The Supplier Grades the Customer: What Pacing the Frontier Means for the Organisations That Buy It
27 September 2026 · 12 minute read
Access to the frontier is now something the supplier allocates, not something the customer buys. The accountability for depending on it stays with the Board.

A senior executive stands in a darkened boardroom looking down through internal glass at an engineering floor lit electric blue, where a smaller glass-walled secure workspace sits apart from the open desks, embodying the Board decision about which work goes to a frontier model and which stays at home. (Image generated by ChatGPT 5) — Nothing Has to Leave the Building: The AI Risk That Is Not a Data Breach
20 September 2026 · 10 minute read
Frontier AI may pass an organisation's know-how to a model its competitors can buy. A Board governs that by deciding which work goes in.

A lone senior engineer studies glowing code on three monitors in a pool of warm light while long rows of empty junior workstations recede into a darkened office at dusk, evoking the widening gap between the volume of AI-generated code needing review and the shrinking supply of people able to debug it. (Image generated by ChatGPT 5) — The Debugging Gap: The Skill Your AI Business Case Assumes You Still Have
29 August 2026 · 10 minute read
AI now writes the code and does the debugging that taught developers to check it. Verification demand rises as the route to building it weakens.

A vast split-flap departures board mounted on the marble wall of a grand institutional concourse, its four rows announcing the EU AI Act's provisions as though they were flights: high-risk systems delayed to December 2027, transparency departed on schedule in August 2026, prohibitions boarding for December 2026, and embedded systems told to wait in the lounge until August 2028, while a lone director in a dark suit stands small beneath the board with a briefcase, reading a timetable that has just changed around him, a visual rendering of the article's argument that the deferral moved some departures and not others, and that the traveller's obligations did not move at all (Image generated by ChatGPT 5.6) — The Deadline Moved: What the EU AI Act Deferral Reveals About Boards
9 August 2026 · 10 minute read
The EU deferred its AI Act high-risk obligations six days before they applied. What a moving deadline reveals about where Board governance is anchored.

A modern open-plan office swallowed by dense fog, where a lone figure sits upright at a desk in the right of the frame, still working, his screens of charts and messages crisp at arm's length while the desks, monitors, and lamps of colleagues dissolve into white haze behind him, a visual rendering of the mental fog the research describes, where the person supervising the machine's output keeps working as clarity recedes, and the cost no business case recorded is paid in human attention (Image generated by ChatGPT 5.6) — The Balancing Item: The AI Oversight Cost Your Business Case Never Priced
26 July 2026 · 10 minute read
Every AI business case counts the hours saved. Almost none counts the oversight hours added, and people are silently absorbing the difference.

A Victorian mill race at dusk seen from the bank, where a broad channel of water pours over a timber weir and is lost in spray and mist lit cold blue among the reeds, while on the right a single well-fitted sluice gate, warm in the light of a lantern hung beside it, directs a narrow measured stream onto the blades of a wooden waterwheel that turns steadily below the lit window of a stone mill workshop, a visual reframe of the Redeployment Dividend, where capacity released by machines is either governed into productive work or drains silently away (Image generated by ChatGPT 5.6) — Governing the Redeployment Dividend: Turning Saved Hours Into Value
19 July 2026 · 10 minute read
AI is saving time almost everywhere. The organisations that gain from it are the ones whose Boards decide what the recovered capacity becomes.

A corner executive office in dark mahogany panelling at night, lit warm on the left by a stained-glass Tiffany lamp beside a porcelain teapot and cup, where a spread of closed leather dossiers fans across a polished desk and one is squared up in the lamplight before an empty high-backed leather chair with a fountain pen resting on it as the chosen bet, the rest angled away into shadow, and cool blue light falls from floor-to-ceiling windows onto a night city skyline and waterfront beyond, a visual reframe of the few bets a chief executive draws from many and answers for, where a machine can lay out the options but only a human can choose (Image generated by ChatGPT 5.5) — AI and the CEO: Choosing the Bets That Matter
21 June 2026 · 12 minute read
AI can build, deliver, and draft, yet the chief executive still chooses and still answers. Accountability for the bets does not move.

A human hand pressed flat against a dark glass barrier at night, reaching toward an illuminated toggle switch mounted in a glass housing on the far side of the glass, out of reach, with a warm-lit city skyline glowing across water beyond, a visual reframe of a control that is visible but held elsewhere, behind a barrier the hand cannot cross (Image generated by ChatGPT 5.4) — The AI Sovereignty Trilemma: When a Frontier Model Vanishes and Reality Bites
14 June 2026 · 10 minute read
The visible cost of sovereignty deters Boards. The hidden cost of the convenient alternative was never shown, and that is the cost 12 June presented.

No articles match that filter. Clear the filter.

I use cookies to understand how my website is used. This data is collected and processed directly by me, not shared with any third parties, and helps me improve my website. See my privacy and cookie policies for more details.

#board-governance

Tagged articles (75)

Filter articles

Humans on the Loop: Governing What an Agent Does, Before It Does It

The Model Did Not Phone Home: Where an Agent's Actions Are Decided

The Principle of Least Intelligence: How Much AI Does the Work Need?

Managers Own the Workflow: Where the AI Business Case Becomes Real

The Supplier Grades the Customer: What Pacing the Frontier Means for the Organisations That Buy It

Nothing Has to Leave the Building: The AI Risk That Is Not a Data Breach

The Debugging Gap: The Skill Your AI Business Case Assumes You Still Have

The Deadline Moved: What the EU AI Act Deferral Reveals About Boards

The Balancing Item: The AI Oversight Cost Your Business Case Never Priced

Governing the Redeployment Dividend: Turning Saved Hours Into Value

AI and the CEO: Choosing the Bets That Matter

The AI Sovereignty Trilemma: When a Frontier Model Vanishes and Reality Bites

Recent