The Workplace AI Policy StarterExpert Edition
One editable AI policy, and everything that makes it work.
One editable AI policy for a small or midsize company, with the acceptable-use list, data handling rules, approval matrix, 30-day rollout plan, and employee acknowledgment form that make it work, built on the NIST AI Risk Management Framework and the vendor, federal, and state rules in force in October 2026.
39 pages · 15 chapters · 3 figures · 3 worksheets · editable Word policy pack included
Available to buyers in the United States. Instant download of the PDF guide and the editable Word policy pack right after checkout, plus both files by email. Payment is processed securely by Stripe, and any sales tax we are required to collect is added at checkout.
Three pages, unedited.
The cover, the contents, and a page from the book exactly as you will receive it. Tap any page to enlarge it.
Written for you if
- Owners and managers of small and midsize US companies where employees already use AI at work without a written rule.
- The person who will own the policy, in operations, HR, or IT, and needs it decided, reviewed, trained, and signed within 30 days.
- Companies that use, or are weighing, software to screen, rank, or score applicants or employees, and need to know which rules come first.
- Teams connecting AI to email, files, customer chats, or code, who want least-privilege access and human approval written into the rules.
Chapter by chapter.
- 01
Why Your Company Needs an AI Policy Now
What Pew Research Center found about chatbot use at work and employer guidance, why the same tool treats your data differently under a personal and a business account, and the four of NIST's twelve generative AI risks that drive most workplace policies.
- 02
The Rules Your Policy Must Respect
The four functions of the NIST AI Risk Management Framework, health data and business associate agreements, hiring rules in states and cities as examples, disclosure for customer-facing AI, FTC rules on AI claims and fake reviews, who owns AI output, and the OWASP Top 10 for LLM applications.
- 03
How to Fill In This Starter
The ten decisions to make before you edit a word, where each one goes in the policy, and three postures, restrictive, balanced, and open, with the clause changes each needs.
- 04
The AI Use Policy (Editable)
The policy itself in fifteen sections, from purpose and approved tools to data handling, human review, disclosure, employment decisions, incidents, training, enforcement, and review, with every company-specific choice marked in [brackets].
- 05
The Acceptable-Use List
Common workplace uses rated green, yellow, or red, each with the condition that applies, ready to post where employees work.
- 06
Data Handling Rules
Four data classes, Public, Internal, Confidential, and Restricted, where each may go, and the rules for minimum necessary use, settings, and uploads.
- 07
The Approval Matrix
Who approves each kind of request, from a yellow use or a new tool to regulated data, customer-facing AI, and any AI in employment decisions, and what must happen before approval.
- 08
Department Addenda
Add-on rules for sales and marketing, HR and recruiting, engineering and IT, finance and legal, and customer support, each tied to the law or risk behind it.
- 09
Reviewing an AI Vendor
What the business pages of OpenAI, Anthropic, Google, and Microsoft say about training, ownership, retention, security, and access, and ten questions to ask any vendor before a tool is approved.
- 10
A Filled-In Example
The policy with every bracket filled in for an invented 40-person design and marketing firm, so you can see a finished version.
- 11
The 30-Day Rollout Plan
Four weeks from decision to signatures, day by day, with an owner for every task.
- 12
The 30-Minute Training Script
A minute-by-minute session with talking points, a ten-question check, and answers to the questions employees ask most in the first week.
- 13
The AI Incident Playbook
Seven common incident types with their default severity, and the first-hour, first-day, and first-week steps, so nobody has to improvise.
- 14
Employee Acknowledgment Form
Seven acknowledgments and a signature block, ready to sign.
- 15
The Quick Reference Vault
The five rules employees must remember, and the key rules and dates from Chapters 1 and 2, to send with the policy.
Three worksheets close the book: an AI tool inventory, a vendor review checklist, and an AI incident log. Answers to every check-yourself question, a glossary, and the complete source list follow. The editable Word policy pack holds Chapters 4 to 14 and the worksheets, with instructions on its first page.
The IKE standard.
Built from primary sources
NIST's AI Risk Management Framework and Generative AI Profile, HHS, the Title VII text, the FTC, the US Copyright Office, OWASP, state and city agencies and codes, a federal court order, and each vendor's own terms. Every factual claim traces to a named, dated source, and a complete source list closes the book.
Editable in Word
The policy, the acceptable-use list, the data rules, the approval matrix, the addenda, the vendor review, the example, the rollout plan, the training script, the incident playbook, the acknowledgment form, and the worksheets come in a Word file, with every company-specific choice marked in [brackets].
Vendor terms, quoted
What the business terms of OpenAI, Anthropic, Google, and Microsoft say about training on your data, quoted from their own pages, read in October 2026. Where a page did not answer a question, the guide says so instead of guessing.
Honest about limits
Rules that were still proposed or in draft are marked unresolved, and the guide says plainly that the template is a starting point to review with employment counsel, not legal advice.
Built to use
3 figures, the critical facts and one action you can finish in under 30 minutes at the end of each chapter in Part I, a filled-in example, three worksheets, and a summary to post or send.
Specs and terms.
- Format
- PDF guide, US Letter, live source links, plus the editable Word (.docx) policy pack
- Length
- 39 pages, 15 chapters
- Inside
- 3 figures, 3 worksheets, a filled-in example, glossary, complete source list
- Sources verified
- October 2026
- Covers
- US employers, mainly small and midsize companies. The template is general: federal rules, with state and city employment laws as examples.
- Price
- $47 USD, one-time purchase
- Delivery
- Instant download of the PDF guide and the editable Word policy pack on a private page right after checkout, plus both files by email
What this is. An educational reference. It is not legal, tax, financial, credit, or investment advice, and it does not create a professional relationship. Rules change; each figure in this edition carries the date it was verified against its official source. For advice on your situation, consult a licensed professional.
Template language and educational content, not legal advice. Have employment counsel review the policy before you adopt it. The purchaser may adapt the templates inside their own organization; reselling or republishing them is not permitted.
Your license. The guide is licensed to the purchaser. The templates in it and in the Word policy pack may be copied, edited, and used inside the purchaser's own organization. Please do not resell or republish them, share the guide outside your organization, upload these files, or use them to train AI models. Terms.
Refunds. Sales are final once your file is delivered, with fair exceptions: if the file is defective, was not delivered, is materially different from this description, or was bought twice by mistake, email [email protected] within 14 days and we will fix it or refund you. Refund policy.
The Workplace AI Policy Starter
39 pages · 15 chapters · 3 figures · 3 worksheets · editable Word policy pack included. $47, instant download.