Discover what exists
Identify AI systems, copilots, embedded AI, shadow AI and autonomous agents.
- AI inventory baseline
- Ownership and use-case mapping
- Data and agent visibility
- Approved versus unapproved use
Personally delivered. Platform enabled.
Fractional AI Governance & Assurance
Senior AI governance and assurance leadership without building a full internal function.
For organisations already using AI that need clear ownership, working controls and defensible evidence, without building a full internal governance team.
Fractional support from £3,500/monthUK-based. Supporting organisations across EMEA.
The governance gap
You probably don’t need another AI policy.
You need to know what AI is in use, who is accountable, where the exposure sits and whether the controls work. Most organisations already have AI activity. Fewer can answer the questions that follow.
What AI are we actually using?
Who owns each system?
What data can it access?
Which agents can act without approval?
Where are the biggest governance gaps?
Which controls are operating?
What evidence could we produce tomorrow?
Who fixes the gaps, and who checks the work?
AI adoption is moving faster than ownership, governance and assurance. The service exists to close that gap.
The service model
Four things I help you get under control.
The work starts with your systems and the decisions your organisation needs to make. It stays focused on ownership, controls and evidence.
Assess what matters
Understand the risks, control gaps, evidence gaps and areas that require attention first.
- Risk and control maturity
- Agentic AI exposure
- Evidence gaps
- Regulatory and framework readiness
Put governance in place
Clarify ownership, decision rights, controls, escalation, remediation and evidence requirements.
- Policies and controls
- Escalation and remediation
- Evidence requirements
- Governance cadence
Prove it works
Provide structured challenge and assurance over whether the programme operates as intended.
- Control and evidence review
- Governance testing
- Board-level visibility
- Remediation oversight
Ways to engage
Start with the level of support the situation requires.
An assessment is a focused, low-risk starting point. A sprint puts the first operating baseline in place. Retained support provides ongoing senior involvement. You do not have to buy all three.
AssessmentSprintOngoing fractional support
AI Governance & Assurance Assessment
From £5,950 fixed
Establish what AI exists, where the major gaps are and what needs to happen next.
- AI inventory baseline
- Risk and governance gap analysis
- Control and evidence observations
- Priority remediation roadmap
- Executive summary
30-Day AI Governance & Assurance Sprint
Fixed scope
Move from findings into an operating governance baseline.
- Inventory refinement and risk classification
- Control assessment and evidence structure
- Ownership and remediation priorities
- Initial governance processes
- Gamut workspace setup
Fractional AI Governance & Assurance
From £3,500/month
Ongoing senior governance and assurance support that works alongside your existing teams.
- Governance, risk and control oversight
- Management and board reporting
- Agentic AI assurance
- Policy and evidence challenge
- Remediation and periodic reviews
Retained support
A fractional model scaled to the responsibility you need covered.
Retained support starts at £3,500/month and scales with how much responsibility you need me to take on. These are service starting points, not unlimited-hours packages.
Advisory
From £3,500/month
Senior guidance, review and governance challenge for internal leaders who retain day-to-day ownership.
Fractional Governance Lead
From £5,000/month
More active ownership of governance, assurance, reporting and remediation oversight.
Embedded Governance & Assurance
From £7,500/month
Deeper involvement for complex environments, multiple AI systems, agentic AI, board engagement or active remediation.
The fractional service provides direct practitioner leadership, judgement, challenge and programme support.
Gamut Assure is licensed separately and maintains the records, assessments, risks, controls, evidence, findings, remediation and reporting.
Why fractional
You may not need an AI governance team yet.
Building an internal capability can be the right long-term decision. It can also take months to recruit the right mix of governance, security, risk, compliance and assurance experience.
You get senior capability now without committing to a permanent senior hire before you know what the long-term function needs to look like.
Fractional support works alongside the CISO, CIO, CTO, risk, compliance, legal and audit teams already carrying responsibility.
Senior AI governance leadership
AI security and agentic AI expertise
Risk, compliance and assurance capability
Governance tooling and evidence management
Operating processes, reporting and remediation
Who you’ll be working with
Arinze Okosieme
Fractional AI Governance & Assurance Lead
You work directly with the senior practitioner responsible for the engagement, not through a hand-off to a junior delivery team.
- 25+ years of cybersecurity experience
- CISSP, CCSP and CCZT
- Trusted AI Safety Expert (TAISE)
- AI governance and agentic AI assurance specialisation
- Creator of Gamut Assure
I work with a limited number of retained clients at any one time so I can remain directly involved in each programme.
Powered by Gamut Assure
The service comes with an operating system.
Gamut Assure gives us a persistent operating record of your AI inventory, assessments, risks, controls, evidence, findings, remediation and reporting. Each governance review starts from a live operating record rather than rebuilding the picture from spreadsheets and slide decks.
AI System Registry
Systems, owners, purpose, suppliers and lifecycle context.
Risk assessments
Traceable risk, applicability and assurance decisions.
Control assessments
Control scope, evidence, tests and conclusions.
Evidence tracking
Proof, quality, gaps and reviewer context.
Findings
Material gaps, owners and decision records.
Remediation
Priorities, accountability and progress.
Audit trail
Reviewable history of governance decisions.
Board reporting
Clear exposure, progress and escalation.
Agentic AI assurance
When AI can act, governance has to go further.
Autonomous and agentic systems do more than generate content. They can use tools, move data and take actions. Governance therefore has to address delegated authority, permissions, runtime conditions, escalation and revocation.
Questions the assurance model must answer
- Who authorised the agent and for what purpose?
- Which tools, systems and data can it access?
- What conditions must remain true while it acts?
- Which actions require human approval or escalation?
- When should its authority be constrained or disappear?
- What evidence explains why an action was allowed?
- Who checks that the controls work?
Framework credibility
Use the frameworks that fit the decision.
Existing policies and governance structures should be used where they work. Frameworks provide depth and traceability, but they do not replace judgement about your actual systems and risks.
NIST AI RMFISO/IEC 42001ISO/IEC 42005EU AI ActGTSAFAgentic Trust FrameworkMAESTRONAGF
Framework support is used for readiness and structured assurance. It does not constitute certification, legal advice, regulatory approval or a guarantee of compliance.
Commercial fit
Best suited to organisations where AI use is already moving.
The strongest fit is where ownership remains unclear, internal teams are stretched, or customers, boards, auditors and regulators are asking for evidence.
Mid-market and compliance-sensitive organisations
Financial services and insurance
Professional services and SaaS businesses
Teams deploying Microsoft Copilot or similar AI
Organisations experimenting with autonomous agents
CISO, CIO, CTO, risk or compliance teams needing additional capacity
Frequently asked questions
What organisations usually ask before we start.
Do we need a full-time AI governance function?
Many organisations do not initially need a full permanent team, but they do need clear ownership and senior oversight. Fractional support establishes that capability while the long-term internal model develops.
Is this the same as a fractional CAIO?
No. This service is focused specifically on governance, risk, controls, assurance and agentic AI rather than broad enterprise AI strategy.
Does this replace our CISO, CIO, CTO, risk or compliance team?
No. It works alongside existing leaders and teams, providing specialist AI governance and assurance capacity.
Is Gamut Assure included?
No. Gamut Assure is separately licensed and provides the operating platform used to maintain the governance programme.
Can you work with our existing frameworks and policies?
Yes, where appropriate. Existing governance structures should be used rather than unnecessarily replaced.
Do you support autonomous or agentic AI?
Yes. Agentic AI assurance is a key area of specialisation, including authority, permissions, runtime controls, human approval, revocation and auditability.
Can you work with organisations outside the UK?
Yes. I am UK-based and support organisations across EMEA.
How quickly can an engagement start?
Start dates depend on current client commitments and the agreed scope. The first call confirms fit, availability and a realistic start date.
Is there a minimum commitment for retained support?
The commitment is agreed as part of the retained scope. It reflects the work required and is confirmed before an engagement begins; it does not imply unlimited hours or availability.
Start with a scoping call
Book a 30-minute call with Arinze.
Tell me briefly what you need help with. We will use the call to decide whether an assessment, sprint or retained engagement is the right next step.