AI readiness: a practical self-assessment

This self-assessment helps leadership teams judge whether they are ready to move an AI use case from idea to production. Score each question 0 (no), 1 (partly) or 2 (yes). The total matters less than the gaps it reveals, because each gap is a predictable reason for a project to stall.

Use case and value

1. Can you name a specific process, owner and user group for your first use case?
2. Have you measured the current baseline: volume, time, errors and cost?
3. Have sponsors agreed the metrics that will decide whether to continue after a pilot?
4. Is there a simpler non-AI alternative, and have you ruled it out for a reason?

Data and content

1. Do you know where the required data or documents live and who owns them?
2. Can engineers get a representative sample within two weeks under appropriate controls?
3. Is the content reasonably current, with superseded versions identifiable?
4. Are access permissions on the source data documented and machine-readable?

Systems and integration

1. Do the systems involved have APIs suitable for reading and, if needed, writing?
2. Is there an agreed cloud environment where AI services can be deployed?
3. Is identity managed centrally so that AI tools can respect user permissions?
4. Do you have a way to monitor services in production today?

Governance and risk

1. Is there an AI usage policy that staff know about?
2. Is there a process to classify AI use cases by risk?
3. Have legal, privacy and security been involved before build rather than at launch?
4. Is it clear who is accountable for the AI system’s outputs in production?

People and operations

1. Will the users who are affected be involved in design and testing?
2. Is there a plan for who maintains the system after launch?
3. Have you budgeted for running costs as well as build costs?
4. Is leadership prepared to stop the initiative if the pilot metrics are not met?

Reading your score