AI governance & assurance · EU AI Act · ISO/IEC 42001
AI governance consulting
AI under control without slowing adoption: every use found, risk tiered, policy and approvals in place, and the evidence to prove it. Mapped to the EU AI Act, ISO/IEC 42001 and NIST AI RMF, at fixed published fees.
Fees published up front
Definition
What is AI governance?
AI governance is the system of policies, decision rights, risk controls and evidence that lets an organisation adopt AI safely and prove it: which uses are approved, who approved them, what risks each one carries, and what would be shown to a board, customer or regulator that asked. Good AI governance speeds adoption up, because approved routes exist and nobody is guessing.
Intelance builds and runs AI governance for organisations in regulated and customer facing environments: inventory, risk tiering, policy, operating model, monthly risk management and independent assurance, mapped to the EU AI Act, ISO/IEC 42001 and NIST AI RMF.
AI governance also stands on data governance: models are only as safe as the data underneath them. Our guide Why data governance is the foundation of safe AI explains the dependency, and the data governance practice covers that side of the house.
Find the AI
Every use inventoried, including the unofficial tools nobody declared.
Tier the risk
Each use classified by impact and exposure, so effort goes where the risk is.
Set the controls
Policy, approval routes and oversight mapped to the EU AI Act and ISO/IEC 42001.
Prove it
Evidence maintained as standard, ready for the board, customer or regulator.
The regulatory clock
The EU AI Act timeline
The Act is already in force and its obligations arrive on fixed dates whether or not organisations feel ready. UK companies are frequently in scope through EU customers, users or group entities.
In force
The EU AI Act entered into force. The clock started for every obligation that follows.
Prohibitions apply
Banned AI practices became unlawful and AI literacy duties began for providers and deployers.
GPAI obligations
Obligations for general purpose AI models began, including transparency and documentation.
Transparency duties
Article 50 transparency duties begin: people told when they interact with AI, synthetic content identifiable, deepfakes disclosed.
High risk obligations
Annex III high risk obligations begin under the June 2026 Omnibus, with embedded systems following in August 2028. Penalties reach €35 million or 7% of global turnover.
August 2026 still matters, whatever you heard about delays
The June 2026 Omnibus moved Annex III high risk obligations to December 2027 and embedded systems to August 2028, but Article 50 transparency duties still begin on 2 August 2026, and evidence takes months to build. The readiness assessment classifies your systems and dates your obligations, from £18,000 + VAT.
Working through it yourself first? Start with the EU AI Act readiness checklist and how to map your AI systems.
Our services
AI governance services
Every service has a published fee and a defined output. Start with the one that matches your situation. They stand alone or run in sequence.
AI Governance QuickScan
Every AI use found, including the unofficial ones, risk tiered with control gaps mapped against the EU AI Act, ISO/IEC 42001 and NIST AI RMF, in two weeks.Best whenThe board has asked what AI is in use and nobody has a defensible answer. FeeFrom £15,000 + VAT · 2 weeks 02AI Governance Office
Policy, risk tiering, approval routes and an operating model built around how your organisation uses AI, designed to run without us.Best whenAdoption is moving faster than governance and you need a working function rather than a document. FeeFrom £30,000 + VAT 03AI Risk-as-a-Service
Monthly use case reviews, control checks and board reporting. Governance that keeps pace with adoption instead of blocking it.Best whenNew AI use cases appear weekly and one off reviews cannot keep up. FeeFrom £8,000 a month 04EU AI Act Readiness Assessment
Your AI systems classified against the Act, obligations mapped to dates, and a remediation plan the board can fund.Best whenThe Act may apply to you and nobody owns the answer. FeeFrom £18,000 + VAT 05ISO/IEC 42001 Implementation
The certifiable AI management system, built on what you already run rather than bolted on beside it.Best whenAn enterprise customer or regulator wants certified AI management rather than verbal assurances. FeeFixed fee, scoped 06AI Assurance Review
Independent evidence that your AI is governed and controlled, written for the board, customer or regulator asking.Best whenSomeone outside the organisation needs independent proof rather than a policy pack. FeeFrom £15,000 + VAT · 2 to 3 weeks 07Supplier AI risk assessment
The AI your suppliers run assessed with the same rigour as your own, because their models are now your risk.Best whenCritical suppliers have quietly become AI companies and nobody has checked. FeeFrom £8,500 + VATNot sure where to start?
Which AI governance service do you need?
Six situations we are usually called into. If one of them sounds like yours, that is the place to start. If none of them fit, book a call and we will point you at the right first step.
How we work
How we deliver AI governance
The same model that runs across every Intelance practice: we assess, we build, and we operate, with independent assurance across all three.
Assess
Find every AI use, tier the risk and map the obligations, before governance is designed.
Services in this stageBuild
Policy, approval routes, an operating model and, where customers demand it, the certifiable standard.
Services in this stageOperate
Monthly use case reviews, control checks and board reporting, run by named senior people.
Services in this stageIndependent assurance runs across all three
When a board, customer or regulator needs proof, the AI Assurance Review provides it independently. If we built your governance, a different and independent team assures it.
Every engagement starts with a thirty minute call and a fixed fee in writing, and ends with evidence you own.
Why Intelance
The standards we hold ourselves to
Mapped to recognised standards
Every deliverable maps to the EU AI Act, ISO/IEC 42001 and NIST AI RMF, so the work holds up when a customer, auditor or regulator reads it.
Written to be shown
Inventories, tiering, policies and reviews are produced as evidence for boards, customers and regulators rather than internal paperwork.
The fee is fixed in writing
Every service on this page has a published fee, confirmed in writing before work starts. Any overrun is our cost and your fee is fixed.
Senior practitioners only
Engagements are led by a senior partner and delivered by senior consultants, with established associates across data governance and cyber where depth is needed.
Structural independence
If we built your governance, a different and independent team assures it. Assurance conclusions follow the evidence and nothing else.
Who delivers this
Every engagement is led by a senior partner. Nothing is offshored and nothing is handed to a junior bench.
AI governance guides
All insights
AI GovernanceAI governance consulting: how to choose a partner in the UK
A practical UK buying guide to AI governance consulting: the five jobs, the four conflicts of interest, the credentials worth checking, the artefacts to demand and the fees to expect.
25 July 2026 · 17 min read Read the guide
AI GovernanceISO 42001 explained: what an AI management system is and whether you need one
25 July 2026 · 8 min read
EU AI ActEU AI Act for UK companies: what applies and when
11 July 2026 · 5 min read
AI GovernanceAI policy template: the two page version people follow
11 July 2026 · 6 min read
AI GovernanceAI governance cost UK: what to budget in 2026
10 July 2026 · 5 min readReady to talk?
Speak to a senior adviser
Thirty minutes on your AI use, your obligations and the right first step. If a board meeting or customer audit is driving this, say so: the QuickScan reports in two weeks.
FAQs
AI governance questions answered
What does AI governance consulting cost in the UK?
With Intelance: the AI Governance QuickScan runs from £15,000 + VAT as a two week fixed fee. Governance build programmes start from £30,000 + VAT. AI Risk-as-a-Service starts at £8,000 a month, with fuller retained governance offices scoped on a call. An independent AI Assurance Review starts at £15,000 + VAT. Every fee is fixed and agreed in writing.
What is in the AI Governance QuickScan?
A two week, senior led assessment: every AI use inventoried including unofficial tools, each use risk tiered, control gaps mapped against the EU AI Act, ISO/IEC 42001 and NIST AI RMF, and a board ready summary of what to fix first. From £15,000 + VAT.
Does the EU AI Act apply to UK companies?
Often, yes. The Act applies to organisations that place AI systems on the EU market, or whose AI outputs are used in the EU, wherever the provider sits. UK companies with EU customers, users or group entities are frequently in scope. The readiness assessment settles the question system by system.
What are the key EU AI Act dates?
The Act entered into force in August 2024. Prohibitions and AI literacy duties applied from February 2025, general purpose AI obligations from August 2025, and Article 50 transparency duties begin on 2 August 2026. The June 2026 Digital Omnibus moved Annex III high risk obligations to 2 December 2027 and embedded systems to 2 August 2028. Penalties reach €35 million or 7% of global turnover.
What is ISO/IEC 42001?
ISO/IEC 42001 is the certifiable international standard for AI management systems: the AI equivalent of ISO 27001 for information security. Certification tells enterprise customers and regulators that AI governance is systematic, audited and maintained rather than improvised.
What is the difference between AI governance and AI assurance?
Governance is running AI safely: policy, approvals, risk tiering and oversight. Assurance is proving it: an independent review that the governance is real and the controls work. We deliver both, structurally separated, and our guide on AI governance vs AI assurance covers the distinction in full.
Will governance slow our AI adoption down?
Done properly, it speeds adoption up. Approved routes and clear tiering mean low risk uses move fast without individual debate, and higher risk uses get scrutiny before they become incidents. The alternative is ungoverned adoption followed by an expensive stop.
Who delivers this on an Intelance AI governance engagement?
Engagements are led by a senior partner and delivered by senior consultants, with established associates across data governance and cyber where specialist depth is needed. Assurance reviews are delivered by an independent team, separate from anything we built. Nothing is offshored.