AI agent for diversity and inclusion managers
Policy Language Inclusion Audit Agent
Make policies inclusive without changing their meaning or legal effect.
What it does
Handbooks and policies can contain wording that excludes people, for example gendered terms, assumptions about family or ability, or unclear accommodation language. When a policy is reviewed, this agent scans the text and flags language by category, with the passage, the concern and a suggested alternative. It tests each suggested change again for meaning and legal accuracy, so an edit does not change a requirement or a legal term. The manager reviews and approves, and legal review is noted where needed. Edge case: a term is defined in law, so the agent leaves it and adds a note.
How it works
Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.
Read the steps as a list
- A policy comes up for review
- Read the policy text
- Scan and flag language by category
- Draft an alternative for each flag
- Does each alternative keep the meaning and legal effect?If not: revise or leave the term and add a note. Back to step 4.
- Compile the change list with reasons
- Manager reviews and approves the listThe agent waits here for your OK.
- Does legal need to review any change?If not: mark those changes and hold them for legal. Back to step 6.
- Produce the clean and tracked versions
- Audit report and edited draft
How it decides
It suggests a change only when the meaning stays the same. Legal and defined terms are left and noted.
- Flag gendered terms, assumptions about family and ability language
- Leave legal terms and defined terms unchanged
- Hold changes that alter eligibility or duties
- Give a reason for every change
Make it yours
Every agent is a starting point. You choose these settings for your own situation.
- Language guide
- Legal terms list
- Policy types
- Who must approve
What keeps you in control
It always asks you first
- Manager approves changes
- Legal approves flagged items
Hard limits
- Never publish a policy
- Never change eligibility without legal
It stops when
- Done: approved edits delivered
- Stop: policy owner rejects changes
Set it up
We guide you through the set-up, step by step
Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.
- One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
- The agent then walks you through connecting your own data, one source at a time
- A downloadable copy with the flow chart, the rules and the full guide