Strategy & Value ยท STR-03
Investment and funding for AI
The organisation has a defined approach to funding AI, covering what is paid for centrally and locally, how long commitments last, and how requests for money are judged.
Why it matters
AI needs some shared spending (platforms, skills, governance) that no single team will fund, and much of its value only appears after more than one budget year. Without a funding approach, promising work stops when a pilot budget runs out, and money goes to whoever asks first rather than to what serves the ambition.
Levels
| 0 Absent | Confirmed absent: the evidence shows nothing is in place. Not known is recorded as insufficient evidence, not 0. |
|---|---|
| 1 Initial | AI is funded from whatever local budgets teams can find; there is no view of total spend. |
| 2 Developing | Some central or ring-fenced funding exists, but the rules for accessing it are unclear, it is year to year only, or it covers pilots and not running costs. |
| 3 Defined | A documented approach to funding AI sets central and local responsibilities, longer commitments where needed and published criteria, and is followed across the organisation. |
| 4 Managed | Total AI spend is known and reported against the funding approach, and funded work is reviewed against what it was funded to deliver. |
| 5 Optimising | Funding allocation is adjusted from evidence of which investments delivered value, and the approach is refined each planning cycle. |
Level 3 criteria
Level 3 is met when every applicable criterion is met. Criteria marked with an overlay apply only to organisations with that characteristic (overlays).
- STR-03.1 A documented funding approach sets out which AI costs are the responsibility of central budgets and which of local budgets.
- STR-03.2 The funding approach provides multi-year commitments where the work needs them.
- STR-03.3 Criteria for judging requests for AI funding are published.
- STR-03.4 Funding decisions across the organisation follow the documented approach.
Assessment questions
- How is AI work funded, and what is paid for centrally versus by individual teams?
- Can work that proves itself in a pilot get funding to run and scale beyond the current year?
- How are competing requests for AI funding judged, and by whom?
- The documented funding approach or investment framework for AI
- Budget papers showing AI allocations, including any multi-year commitments
- Records of recent funding decisions against the stated criteria
How to reach Level 3
- Decide which AI costs are paid centrally (for example shared platforms, licences and governance) and which are paid by individual teams.
- Identify which AI commitments need funding over more than one year, such as platforms, contracts and skills, and agree how they are protected.
- Write and publish the criteria used to judge funding requests, linked to the value case (STR-04) and prioritisation criteria (POR-02).
- Document the funding approach, get it approved by finance and leadership, and apply it to every AI funding request from then on.
Common pitfalls
- Funding AI only through one-year pilot budgets, so successful uses have no money to run once the pilot ends.
- Publishing criteria but letting requests bypass them through local budgets.
Starter artefacts
- AI funding approach note setting out central and local responsibilities
- Published funding request criteria and scoring sheet
Ownership and sequence
- Accountable
- executive sponsor
- Contributors
- ai governance lead, business owner, delivery lead
- Depends on
- STR-01
- Roadmap step
- 3
- Lifecycle
- strategy, design
Cross-references
| ISO/IEC 42001 | 5.1 partial; 7.1 direct |
|---|---|
| MITRE AI Maturity Model | P2.1 direct |
| NIST AI RMF 1.0 | MANAGE 2.1 partial |
| UNESCO AI Maturity Framework | 1.1 partial; 1.5 partial; 3.2 partial |
direct: this control at Level 3 meets the item. partial: it covers part of it.