AI Due Diligence
What stands behind the AI promise?
Before an investment or acquisition, what matters is what a company can actually do. We make AI capabilities, dependencies and untested assumptions visible for your decision.
The short answer
AI SHIFT’s AI due diligence examines a company’s AI capabilities and value potential. It delivers a clear AI Impact Score based on our assessment model: the foundations are AI readiness across data, IT infrastructure, the software stack and organisation, and the gap to the maximum score. The report connects the score with evidence, capability gaps and prioritised areas for action.
When AI is part of the investment thesis.
A convincing demo does not establish whether a solution works in daily operations, has sustainable economics or can evolve without individual key people. AI due diligence addresses that gap.
- Is AI already in operation, or still in the pitch deck?
- Which data and systems support the solution – and who controls them?
- What will it cost to realise the potential being promised?
Assessment criteria
Four perspectives on the target.
Data & evidence
Data availability, quality and provenance, alongside demonstrable usage and results. Missing evidence remains visible as an open issue.
Architecture & dependencies
Model and vendor dependencies, interfaces, operational readiness and technical debt. A demo is distinguished from a production system.
Organisation & ownership
Capabilities, responsibilities and key-person risks. Who can operate, evaluate and develop the system further?
Economics & potential
Benefit assumptions, running costs and follow-on investment. Potential is tied to prerequisites rather than treated as assured returns.
Deliverables
A basis for the investment committee.
- An AI Impact Score based on our assessment model, derived from AI readiness and the gap to the maximum score
- Prioritised risks, dependencies and evidence gaps
- Value opportunities with assumptions, prerequisites and investment needs
- Recommendations for further investigation and the next roadmap
Approach
From scope to decision.
Define the question
We clarify the investment thesis, available documents, contacts and timeframe. These define the review scope.
Review the evidence
System reviews and discussions with responsible people connect technical findings with actual operations. Access gaps are documented.
Assess the findings
We assess AI readiness and derive the AI Impact Score from the gap to the maximum score. Findings and capability gaps make the assessment traceable.
Prepare the decision
A report and findings discussion translate the evidence into implications for your decision and next review steps.
Common questions
How does AI due diligence differ from technical due diligence?
Technical due diligence examines a company’s overall technical substance. AI due diligence goes deeper into AI-specific questions: data foundations, models, evaluation, vendor dependencies, operations and the economic viability of AI claims. The reviews should be coordinated.
Who is this review for?
Investors, family offices and acquirers whose decisions depend on AI capabilities or AI-driven changes to a business model. Even when a target has no AI product, understanding opportunities and prerequisites can matter before an investment.
Do we receive an AI Impact Score, and how is it determined?
Yes. You receive a clear AI Impact Score based on our assessment model. The starting point is AI readiness: we assess data, IT infrastructure, the software stack and organisation using our existing readiness system. We derive the AI Impact Score from that assessment and the gap to the maximum score. It makes the remaining AI potential visible within the assessment model. The report explains the underlying findings, gaps to the target state and resulting areas for action.
How long does the review take, and what does it cost?
That depends on the system landscape, scope, available access and transaction timeline. We clarify the scope in an introductory conversation and then prepare a suitable proposal. A fixed duration or price without that context would not be reliable.
Let us clarify what your decision needs.
Bring your investment thesis and open AI questions. In an introduction, we establish which review is useful and feasible with the access available.