Use cases / People and insight
AI HR
Help employees and HR teams, with people decisions left to people.
01
Problem
HR answers repetitive questions while handling the most sensitive data in the organization. Automated decisions about people carry legal and ethical risk.
02
AI capability
Answer policy questions, guide onboarding, draft job descriptions and summarize candidate materials for a recruiter.
03
Data
The agent works from the context you connect, not from the open internet by default.
- HR policies and handbooks
- Onboarding checklists
- Job requisitions
- Employee records (tightly scoped)
04
Agent
HR Agent
Answers policy questions and prepares materials; never decides on individuals.
05
Workflow
- 1. Question or request
- 2. Check requester permissions
- 3. Retrieve policy
- 4. Answer or draft
- 5. Escalate sensitive cases
- 6. Log access
06
Governance
Governance runs inside the agent at runtime: policy changes what it can actually do.
Can
- Answer from published policy
- Draft job descriptions
- Guide onboarding steps
- Summarize materials for a recruiter
Cannot
- Rank, score or reject candidates alone
- Reveal another employee data
- Decide on pay, discipline or termination
- Retain sensitive data beyond policy
Requires approval
- Any access to an individual record
- Messages about employment status
- Candidate shortlists
Records
Agent identity, Data accessed, Model used, Output, Tools called, Policy applied, Decision, Approval, Action, Outcome.
Suggested starting autonomy: L1 Assist
Start at L1: decisions about people stay human; the agent assists with information and drafts.
07
Outcome
What you measure, so the agent earns more autonomy on evidence:
- HR time on routine questions
- Onboarding task completion time
- Escalations handled
- Employee-record access events reviewed
Questions
Does it make hiring decisions?
No. It prepares materials; people make and own decisions about candidates and employees.
Who can see HR data through it?
Only users already permitted to, and every access is recorded.