Lesson 46 of 1550
Expense Policy Assistants: 'Can I Expense This?'
Every finance team gets the same question 50 times a week. A policy-grounded assistant answers consistently and reduces compliance risk.
Lesson map
What this lesson covers
Learning path
The main moves in order
- 1The policy paradox
- 2policy grounding
- 3compliance
- 4natural language
Concept cluster
Terms to connect while reading
Section 1
The policy paradox
Expense policies are written for auditors, not for employees. An employee asking 'can I expense business-class on a 14-hour flight' shouldn't have to read 30 pages of legalese. A grounded policy assistant translates the legalese into a yes/no/depends answer with citations.
What 'grounded' means here
- 1Answers cite the exact policy section
- 2Conflicts between policy and country-specific addendums are surfaced, not hidden
- 3Anything not in the policy returns 'not addressed — please ask Finance' rather than a guess
- 4Edge cases get logged as candidate policy updates
The audit trail is the product
If an auditor asks why an employee thought a $400 dinner was reimbursable, 'the assistant said yes' is an answer — IF you logged the assistant's exact response, the policy version it cited, and the timestamp. That logging is non-negotiable.
Key terms in this lesson
The big idea: AI for compliance is about consistency and audit, not about removing human approval.
End-of-lesson quiz
Check what stuck
15 questions · Score saves to your progress.
Tutor
Curious about “Expense Policy Assistants: 'Can I Expense This?'”?
Ask anything about this lesson. I’ll answer using just what you’re reading — short, friendly, grounded.
Progress saved locally in this browser. Sign in to sync across devices.
Related lessons
Keep going
Adults & Professionals · 40 min
SOP Automation: Turning Tribal Knowledge Into Prompted Workflows
Standard Operating Procedures live in PDFs nobody reads. An LLM can compile them into living, prompt-driven checklists that adapt to context.
Adults & Professionals · 10 min
Internal-Doc Q&A Bots: Beyond The Wiki Search Box
Most companies have years of policies, runbooks, and onboarding docs nobody can find. A grounded Q&A bot makes that knowledge addressable — when built carefully. A grounded LLM bot reads the question, retrieves the relevant passages, and answers with citations.
Adults & Professionals · 8 min
Meeting Summarization: Beyond The Generic Recap
Meeting recap tools are everywhere. Most produce summaries that nobody reads. Here's how to design summaries that drive action. Establish a meeting-by-meeting consent norm — 'this meeting is being summarized by AI' — and respect opt-outs by turning the bot off, not by hoping it won't notice.