Tool 01 / 20–30 minutes
Assess an AI evidence claim.
Separate a headline claim from what the evidence actually establishes. Identify the questions that could change your decision.
Your output: An evidence assessment and a prioritised request for further evidence.
When to use this tool
For a specific claim about an AI system or intervention. Record the study and version you reviewed. A technical performance result alone does not establish effectiveness, fairness or suitability in your service.
Loading your workspace…
Read the evidence / Reviewed 18 October 2026
Sources & scope.
AIPEX’s prompts and worked approaches are our own synthesis. Use the linked material to examine the method and its limits.
- HM Treasury
The Magenta Book (2026) ↗
Evaluation design, impact evaluation and the counterfactual
Use to distinguish observed change from effects attributable to an intervention. Evaluation design must fit the question and context.
- UK Government
Guidance on the impact evaluation of AI interventions ↗
Evaluating AI interventions in their operational setting
Helps frame evaluation of the intervention and its workflow, alongside technical model testing.
- Mitchell et al.
Model Cards for Model Reporting (2019) ↗
Intended use, evaluation data, metrics and subgroup performance
A reporting proposal, useful for asking what an evaluation covers and leaves out. A model card is not independent validation.
- Information Commissioner’s Office
AI in Recruitment Outcomes Report (2024) ↗
Report pp. 5, 9–10 (scope); p. 20 (data separation); pp. 24–25 (Organisation A and bias monitoring)
Historical findings from selected consensual audits conducted August 2023–May 2024. Generative AI was outside scope. Findings do not estimate prevalence across the recruitment market or state current law.
Go further.
Develop the underlying skills through the related course, or examine another case.