Use case · People · Flagship
Rewrite one role with clear AI / review / human-only lines
For businesses that added AI tools without rewriting responsibility: pick one role, mark every task as AI does · Human checks · Human only, set hard limits, and drill one scenario before relying on it.
Baseline
Tools exist but nobody owns AI-assisted errors, so adoption stalls. Pick the role where AI already touches the work most.
Sample output
One-page role card: 14 tasks — 5 “AI does”, 6 “Human checks”, 3 “Human only” — plus written limits (“AI never sends pricing over $2,000 without approval”).
Illustrative example. Not a live customer result.
Inputs
- The current task list for one role
- Money and risk limits you are not willing to cross
- A backup person for each AI-handled task
What AI prepares
- Draft the first pass of the task split for the owner to edit
- Draft the scenario drill script from a real past incident
- Summarise drill results into proposed rule updates
A person approves
- The manager owns the final task split and all limits
- AI never decides hiring, pay, performance or dismissal
- Every AI-handled task has a named backup person