Data
Whether the information AI needs is accessible, trustworthy and allowed to be used.
Where does the information for your key workflows live?
Why it matters: AI can only use information it can reach.
A first step: Map the systems behind your top workflow and move key documents out of inboxes and personal drives.
How much do you trust that data?
Why it matters: AI outputs inherit the errors in the data behind them.
A first step: Name an owner for your most important data set and fix the errors that matter most for the first use case.
Do you know which of your data is personal or confidential?
Why it matters: That decides what can safely go into an AI tool.
A first step: Sort your data into simple classes, such as public, internal, confidential and personal, and set AI rules for each.