AI Apprenticeship Units Strategy 2026: Build the Right Portfolio

Last reviewed: 15 July 2026 against the approved Skills England portfolio and current unit funding rules.

The strategic opportunity is not simply to buy three short courses. It is to place the right leaders at the right decision point: setting direction, choosing and governing systems, or turning a decision into safe organisational change. A portfolio designed around job titles or a default three-unit bundle wastes time and can create duplicate-funding risk.

Treat the Units as Three Capability Layers

These layers form a useful operating model, but they do not prove that one person needs all three. In a large organisation, strategy, procurement governance and transformation may sit with different leaders. In a smaller organisation, one person may span the lifecycle and still have enough prior learning to make one of the units unnecessary.

Map Units to AI Maturity

• Exploring: the organisation has scattered experimentation but no agreed priorities. AU0009 can help an accountable leader turn ideas into an evidence-led direction.
• Selecting: use cases exist and leaders must compare tools, suppliers and deployment models. AU0010 develops procurement, risk and governance capability.
• Deploying: approved systems are moving into workflows. AU0011 focuses on change, workforce impact, incident readiness and implementation risk.
• Scaling: several systems are live and controls must work consistently. AU0010 may address governance gaps; AU0011 may address monitoring and transformation gaps. Initial assessment decides which.

Maturity should be assessed per business area, not only enterprise-wide. A bank might be scaling AI in fraud operations while still exploring it in HR. The leaders in those functions may need different units even under the same corporate AI policy.

Build Cohorts Around Shared Decisions

Cohorts work best when participants face comparable decisions and can use a common language. Examples include:

• functional leaders building an enterprise use-case portfolio on AU0009;
• technology, procurement, data, security, legal and risk leaders working through vendor and governance decisions on AU0010; or
• operations, HR, transformation and service leaders planning implementation and workforce transition on AU0011.

A mixed cohort can be valuable, but every learner must meet the role and funding rules. Skills England requires an employed learner aged 19 or over in a leadership position with autonomy to deliver technological change and inform investment decisions. General awareness populations belong on a different learning route.

Use five gates before approving a place

1. Role: does the employee make or materially shape the decisions covered?
2. Need: which published knowledge and skills are genuinely missing?
3. Prior learning: would existing capability overlap with 50% or more of planned unit hours?
4. Application: is there a safe, current workplace problem on which to practise?
5. Validation: can a line manager observe and validate successful application?

Sequence by Business Gates, Not Calendar Convenience

If one leader has eligible needs across the lifecycle, connect the units to real milestones:

1. complete AU0009 before approving a strategic use-case portfolio;
2. complete AU0010 before a material vendor or governance decision;
3. complete AU0011 before or during controlled implementation, while change and monitoring plans can still be shaped.

Do not compress three units into an old 80-hour syllabus. Each current unit requires at least 30 eligible delivery hours and its own valid skills test. Nor should all three be sold as a £2,000 programme: the published maximum is £750 per unit, so three separately eligible units have a maximum combined government contribution of £2,250. The agreed provider price and employer contribution can differ.

Measure Decisions, Controls and Change

Programme completions, attendance and skills-test passes remain essential evidence. They should sit beside operational outcomes, not substitute for them.

A 90-Day Portfolio Roadmap

1. Days 1–15: map decisions. Identify where AI direction, selection, governance and implementation decisions sit across the organisation.
2. Days 16–30: assess roles and gaps. Apply eligibility and prior-learning gates; do not begin with a nominee list.
3. Days 31–45: select units and provider. Use exact published outcomes, hours, assessment and funding figures.
4. Days 46–60: define application. Agree safe workplace outputs, data boundaries, line-manager validation and business measures.
5. Days 61–90: launch the first targeted cohort. Start with the business gate that is both urgent and ready, then review evidence before scaling.

Providers needing delivery-hour, assessment and evidence detail should use the current provider guide. Employers evaluating suppliers should use the AI leadership unit buyer's guide.