02 · New for 2026 · Leadership

AI without the hype. Governance without the friction.

Most boards don't need an AI strategy. They need a position. We help yours write one that survives audit, regulator, and the next CFO budget review. And identifies the two or three places AI actually deserves your capital this year.

Book a 30-min intake → Read the insights → All eleven services

Three concentric rings: HITL, HOTL, and HIC. The three modes of human oversight over AI.

Why us

Two reasons. The rest is detail.

01 · Practice over theory

We run the same machinery on ourselves.

We don't talk about AI in theory. We run an internal R&D programme: gated pilot lifecycle, versioned prompts, evaluation suites, eval gates before any prompt change, structured logs, AWS budget alerts. The same governance machinery we install for clients is the machinery we run on ourselves.

02 · CIO, not specialist

The position has to fit your operating model.

We're CIOs, not AI specialists. The position has to fit your IT operating model, your security posture, your compliance trajectory, and your CFO's view of OPEX. AI consultancies that haven't run an IT P&L tend to miss at least one of those.

Three modes of human oversight

Where the human sits. Decides what the system can do.

Every AI use case in your business sits in one of three modes. The board's AI position must name which mode applies where. Mixing them up is the most common governance failure we see.

HITL · Highest control

Human-in-the-loop

The human sits inside the decision chain. The system cannot act without explicit human approval per action.

Use it for: financial commitments, HR decisions, customer contracts, security actions.

Trade-off: highest control. Lowest speed.

HOTL · Balanced

Human-on-the-loop

The human sits above the chain as supervisor. The system acts autonomously. The human monitors actively and can intervene or stop at any moment.

Use it for: operational automation, monitoring alerts, routine integration flows.

Trade-off: balance between speed and control.

HIC · Highest speed

Human-in-command

The human sits in front of the chain. Sets the strategy, policy, boundaries, and kill-switches. Does not intervene operationally. The system runs within those frames; review happens after the fact via audit trails.

Use it for: high-volume, low-risk automated processes.

Trade-off: highest speed. Requires strong post-hoc governance.

The problem boards have

Three internal AI initiatives. Zero owners senior enough to kill them.

Pressure from outside

Competitors talk about AI. Customers ask what your AI position is. Auditors want to know which controls apply. The EU AI Act lands in stages.

Confusion inside

Three internal initiatives running in parallel. None with an owner senior enough to kill them. None with an evaluation framework anyone trusts.

Vendors quoting

Tools you haven't decided you need. Demos that drove timeline. Procurement is tracking spend you can't yet justify on the board pre-read.

What this engagement is

A board-ready AI position. In six to ten weeks. One document.

We don't sell pilots. We don't sell prompts. We don't sell vendor relationships. We sell the position itself, and the discipline behind it.

How it works

Six gates over six to ten weeks.

No invented frameworks. The same gates we use in our own R&D programme.

Gate 01 · Week 1

Inventory and alignment

What AI is already running in your business. Sanctioned, shadow, or vendor-embedded. Who owns it. What it's costing.

Gate 02 · Week 2–3

Risk and regulation map

EU AI Act applicability per use case. NIS2, GDPR, sector-specific rules. Customer contract obligations. Insurance position.

Gate 03 · Week 4–5

The two or three bets

Where AI actually deserves capital this year. Justified against P&L, not against competitor announcements. Everything else: sunset or defer.

Gate 04 · Week 6–7

Governance machinery

Pilot lifecycle gates, evaluation framework, prompt versioning, cost monitoring, human-in-the-loop policy.

Gate 05 · Week 8–9

Board document

The single document the board adopts. Written so non-technical directors can defend it in front of auditors and shareholders.

Gate 06 · Week 10

Operating handover

To the executive accountable internally. We don't stay to "support adoption". The position should be operable by your team on day 71.

Deliverables

Six artefacts. The board adopts one. Your team operates the rest.

The Position document

Single source of truth, board-adopted, audit-ready. Six to ten pages. Defensible to auditors and shareholders.

Use-case ledger

Every AI initiative ranked by risk, value, and regulatory fit. Quarterly review template included.

Governance kit

Pilot lifecycle gates, evaluation template, prompt-versioning standard, cost-monitoring set-up.

Risk and regulation map

EU AI Act tier per use case. NIS2 / GDPR / sector overlay. Insurance position.

Vendor and contract review

What you've already signed for. Where you have room to renegotiate. Where you're locked in.

Operating handover pack

For the executive who runs it after we leave. Working binder, not a deck.

Common questions

Six things buyers ask before the call.

How is this different from a Big-Four AI strategy engagement?

Smaller, faster, and we do the writing ourselves. A Big-Four engagement gives you a 60-page deck and a senior partner who shows up at the readout. We give you a 6 to 10 page Position document the board adopts, plus the governance kit you need to operate it.

Do you build the AI tools?

No. We define the position, the controls, and the governance. We don't take vendor commissions, we don't bundle implementation, and we don't have a pet platform we're trying to get you onto. This is by design.

What about the EU AI Act?

Built into Gate 02. Every use case is classified against the Act's risk tiers, with the obligations that apply. Where the Act is silent or still being clarified, we say so explicitly.

We've already started AI work. Do we have to pause?

No. The first thing we do is inventory what's running, sanctioned and otherwise. Some of it will pass review. Some won't. The Position document tells you which is which.

Who internally needs to be involved?

The CEO or board chair to sponsor it. The CFO and General Counsel for the regulatory and budget conversations. The CIO/CTO and CISO for the operational reality. Plus whoever inside the business is closest to the use cases under review. Total time per executive: 4 to 6 hours over the engagement.

What does it cost?

Fixed scope, fixed fee. We share the price after the intake call, once we know the size of your AI footprint, the regulatory complexity, and how many use cases need ranking.

Where to from here

Bring the AI conversation your board keeps having and not finishing.

Leave with a six-week plan to finish it.

Book the call → All eleven services