The State of AI Apprenticeships 2026

AI apprenticeships moved from a niche idea to a stated government priority in under two years. At the same time, the funding system underneath them was rebuilt. This report pulls the two threads together: what is actually funded in 2026, what changed, what the skills-gap data says about demand, and what a good provider outcome looks like. It is written for employers, L&D leads and journalists, and you are welcome to cite any figure here with attribution.

1. AI apprenticeships are now a policy priority, not a fringe

The UK government's AI Opportunities Action Plan, published in early 2025, named apprenticeships as a core route for closing the country's AI skills gap. That signal matters: it moved AI training from a "nice to have" into the same policy conversation as the apprenticeship levy itself. In practice, demand has concentrated around two things, AI literacy for the broad workforce, and AI adoption capability for managers and operational leaders, rather than only deep machine-learning engineering.

The provision has followed. Alongside established digital and data standards, 2026 saw the arrival of short, stackable Level 5 AI Leadership units aimed squarely at senior leaders, a format that did not exist a year earlier.

2. The 2026 funding reset: the Growth & Skills Levy

From 1 April 2026 the Apprenticeship Levy became the Growth & Skills Levy. The rate and core mechanism are unchanged (0.5% of payroll over £3m), but the rules around spending changed materially. The five changes that matter most:

• Levy expiry halved. Funds deposited from April 2026 now expire after 12 months, not 24. This forces quarterly planning rather than annual.
• The 10% top-up ends on 1 August 2026, removing the automatic bonus on contributions.
• Co-investment rises to 25% (from 5%) for levy-paying employers who exhaust their pot and want to keep funding apprenticeships. Non-levy SMEs are unaffected and still pay 5%.
• Short apprenticeship units (30 to 140 hours) became levy-eligible for up to 50% of the pot, opening a route to rapid, modular upskilling, including AI leadership.
• 16 standards are defunded from 1 September 2026 (more on this below).

For SMEs the direction is generous: under-25 apprentices are 100% funded from October 2026, with a £2,000 cash incentive on top. The net effect is a system that rewards employers who plan deliberately and penalises those who let levy sit.

3. What is being defunded, and what is replacing it

The 16 defunded standards land hardest on leadership and management. Team Leader (L3), Operations Manager (L5), Coaching Professional (L5) and Chartered Manager (L6) are all affected. Existing apprentices are protected to completion, and new starts are allowed until 17 December 2026, but the long-term funded route for those exact standards closes.

The market response has been to move buyers onto adjacent, still-funded routes that keep a recognised management qualification embedded. For example, the defunded Team Leader L3 maps to an AI-embedded leadership apprenticeship that retains a CMI Level 3 credential, so learners keep the qualification while gaining AI-adoption skills the old standard never covered. We cover the full list and alternatives in Apprenticeship Defunding 2026.

4. The demand driver: an AI adoption gap, not just a hiring gap

The story underneath the policy is an adoption gap. Most organisations now have AI tools deployed (Copilot in seat licences, ChatGPT in pockets) but uneven literacy and almost no governance. The bottleneck has shifted from "do we have the tools" to "can our people actually use them, safely, in the real workflow." That is a training problem, and it is why employer demand has skewed towards practical, no-code AI literacy for the existing workforce rather than only specialist engineering hires.

This is also why the apprenticeship route fits: it funds sustained, workplace-applied training (a real project shipped, not a one-day course) at a moment when employers are nervous about both the skills gap and the risk of unmanaged AI use.

5. What good outcomes look like

Outcome quality varies enormously between providers, and End-Point Assessment (EPA) results are the cleanest comparison available. The headline metric is the distinction rate. As a published benchmark, TESS Group reports a 72% distinction rate at EPA against a national average of around a third, with a 98% overall pass rate, independently verified by its EPA partners and refreshed quarterly.

The lesson for buyers is to ask every provider for the same two numbers, distinction rate and overall pass rate, by standard, and to treat a provider that cannot produce them as a red flag. From 2026, Skills England approval (formerly RoATP) is also becoming mandatory to deliver levy-funded training, which gives employers a second hard filter.

6. What employers should do before September 2026

• Audit your levy now. Identify funds expiring under both the old 24-month and new 12-month rules (the system uses first-in-first-out).
• Check your exposure to the 16 defunded standards. If you run Team Leader, Operations Manager, Coaching Professional or Chartered Manager, plan the switch before the December enrolment cut-off.
• Use units for urgent gaps. Short AI leadership units can address adoption fast while you reserve levy for deeper apprenticeships.
• Benchmark providers on outcomes, not price. Ask for distinction and pass rates by standard, and confirm Skills England approval.

Methodology & sources

This report synthesises current UK apprenticeship policy with TESS Group's own delivery data. Policy and funding facts follow official GOV.UK guidance on the Growth & Skills Levy and the AI Opportunities Action Plan. Outcome figures (72% distinction, 98% pass) are TESS Group's own, independently verified by its EPA partners and refreshed quarterly. National comparison figures are presented as TESS Group's stated benchmark. Where a figure is TESS's own rather than a national statistic, it is attributed in the text.