data-readiness

AI Readiness and Data Foundations

Public guide to checking whether documents, data and ownership are ready for reliable AI workflows.

Many AI projects fail because the workflow is unclear, the data is not owned or the documentation is not safe to expose. AISYSTEMS readiness work checks the inputs before building more automation.

Readiness questions

• Which documents are public, internal or restricted
• Which data is accurate enough for automated assistance
• Which actions require a person before execution
• Which systems can be read but not written
• Which logs or evidence are needed for review

Practical first slice

The first slice should usually be read-only: collect approved sources, create a small index, define refusal rules and test whether answers are grounded in the right material.

Output

The output is a readiness map with source classes, gaps, risks and the smallest safe workflow that can be tested.