AI Governance & Assurance
AI Impact Assessment
Assess a specific AI use case, its potential impacts and the safeguards and evidence needed for an accountable decision.
A structured assessment of one AI use case — distinct from an organisation-wide governance review.
Not sure this is the right service? Find the Right Service
When this service fits
For a planned or live AI use case that affects people, decisions or regulated outcomes and needs a defensible go, condition or stop decision.
A new AI use case affects customers, patients or staff
A supplier is introducing AI into a service you rely on
Leadership needs a defensible basis to approve or decline
A DPIA or clinical safety review raised AI-specific questions
An existing AI use has changed in purpose or scale
Procurement or a regulator is asking how the use case was assessed
The service
One use case, examined properly
The assessment looks at the intended purpose, the people affected, the data involved, supplier dependencies, human oversight and the relevant legal, privacy, safety, security and operational risks — then sets out the safeguards and evidence needed before a decision.
It connects with, but does not replace, a DPIA, clinical safety assessment (DCB0129/DCB0160), cyber review or any applicable regulatory assessment. Where one of those is required, it is identified and scoped on its own terms.
Management remains responsible for the deployment decision and for accepting residual risk.
What changes for your organisation
Outcomes the service is built to deliver
Use case and purpose
What the AI does, for whom and why, within agreed boundaries.
Affected people
Who could be affected, how, and how they could raise concerns.
Supplier dependencies
What the supplier provides, claims and evidences.
Oversight
How humans monitor, intervene in and override outputs.
Relevant risks
Legal, privacy, safety, security, fairness and operational risks proportionate to the use.
Safeguards and evidence
Controls needed, evidence required and conditions for a decision.
How we deliver
Assess → Build → Manage → Assure → Improve
Assess
Confirm the use case, scope and evidence available.
Build
Assess impacts and define safeguards and conditions.
Manage
Optional — ongoing review is scoped separately if wanted.
Assure
Present a decision record to management.
Improve
Reassess when purpose, supplier or scale changes.
Improve feeds the next Assess cycle, so evidence stays current.
What you receive
The working records this service produces
Illustrative examples of the registers, records and executive reporting that may support the engagement.
- AI systems registered
- 16 ▲ 3 this period
- Impact assessments open / closed
- 3 / 9 ▲ 2 closed
- Systems awaiting approval
- 2 No change
- AI issues under review
- 1 No change
AI risk by level
Controls evidenced
76%
- Model reviews due
- 3
- Overdue actions
- 2
- Matters for escalation
- 1
Illustrative output
AI Governance Dashboard
Ongoing oversight
A recurring view of AI systems, assessments and controls, so AI risk is seen and escalated before it becomes exposure.
What it helps you see
- AI system inventory and movement
- Impact assessments and approvals
- AI risk by level
- Reviews, actions and escalations
Likely format
Dashboard / reporting view · Supporting registers and evidence records · Executive PDF summary where agreed
| Impact | Who | Likelihood |
|---|---|---|
| Incorrect output | Service users | Possible |
| Data reuse | Customers | Unlikely |
Illustrative output
AI Impact Assessment Record
Assessment
The structured assessment of the use case.
What it helps you see
- Purpose and context
- Affected people
- Impacts and risks
- Safeguards
Likely format
| Decision | Condition | Owner |
|---|---|---|
| Proceed with conditions | Supplier evidence | Owner A |
Illustrative output
Decision & Conditions Record
Decision
What management is asked to decide, and on what conditions.
What it helps you see
- Recommendation
- Conditions
- Residual risk
- Decision owner
Likely format
Illustrative structure — not a client document. Exact outputs and formats depend on the agreed scope.
Scope boundaries
What is included, and what is scoped separately
Included in the core service
- Use-case scoping
- Impact and risk assessment
- Review of supplier evidence within scope
- Safeguards and conditions
- Decision record for management
- Identification of related assessments required
Separately scoped where required
- DPIAs, clinical safety cases or regulatory submissions themselves
- Technical model testing or red teaming
- Formal legal opinions
- Organisation-wide governance assessment
- Implementing safeguards
Scope, deliverables and assumptions are agreed in writing before work begins.
Relevant evidence
Relevant governance & assurance capability
This evidence shows related HealthTech, governance and assurance capability. It is not presented as a completed AI impact assessment unless the case says so.
View the evidence: Clini-HubNamed client · Technology / SaaS / RegTech
Clini-Hub
Embedded governance and clinical-safety support since start-up
NHS AI Toolkit, DSPT, DTAC and clinical-safety governance from start-up
View the evidence: A FTSE 100 organisationAnonymised engagement · Technology / SaaS / RegTech
A FTSE 100 organisation
Three-stage governance and audit-readiness plan
Audit-ready UK policies ahead of a third-party maturity assessment
View the evidence: AIGNamed client · Financial Services
AIG
Supplier data-governance audit
Nationwide supplier audit
Your IG-Smart team
Specialist expertise, coordinated around your requirement
Subject-matter expertise is paired with a clear client and programme contact from initial scoping through delivery.


Client & programme contact
Julia Andrade
Coordinates scope, practitioners, delivery and stakeholder communication.
View profile
Buyer decisions
Questions before you engage
How is this different from an AI governance assessment?
Does it replace a DPIA?
Who decides whether the AI is deployed?
Can you assess a supplier's AI?
Investment
Scoped to requirement
Scope, deliverables, assumptions and fees are agreed in writing before work begins.
The fee depends on
- Complexity and impact of the use case
- Affected people and data
- Supplier evidence available
- Related assessments to coordinate
- Timescale for the decision
Procurement or supplier-assurance review?
Visit our Trust CentreReady to move forward?
Need a defensible decision on a specific AI use case?
Not sure which service applies? Find the Right Service
Still defining your requirement?
Discuss an AI Use-Case Assessment
Discuss the requirement, risk, scope and the right engagement model with an experienced practitioner.
Discuss an AI Use-Case AssessmentHave a defined scope, tender or RFP?
Submit an AI Impact Assessment Requirement
Share a defined requirement, RFP, tender, statement of work or existing scope for senior review.
Submit an AI Impact Assessment RequirementProcurement or supplier assurance
Prepare for Procurement Review
Access company, security and assurance information for supplier review, with controlled evidence available on request.
Open Trust Centre

