AI Policy Starter Kit for Small Business

AI USE POLICY
[Company Name]
Last updated: [Date]

SECTION 1: PURPOSE AND SCOPE

This policy governs the use of artificial intelligence (AI) tools at [Company Name]. It applies to all employees, contractors, and partners who use AI tools in the course of their work.

The goal is to enable productive use of AI while protecting our clients, our team, and our reputation.


SECTION 2: APPROVED USES

AI tools may be used for the following:
- Drafting content, with human review before any external use.
- Brainstorming and ideation.
- Data analysis, with verification against source data.
- Code generation, with testing and review before deployment.
- Summarizing publicly available information.
- Translation.

AI tools must not be the final decision-maker for any client-facing, financial, legal, or HR decision.


SECTION 3: PROHIBITED USES

- Do not enter confidential client data, trade secrets, or personally identifiable information (PII) into public AI tools unless the tool has a documented data protection agreement.
- Do not use AI to generate content that is presented as human-created without disclosure, where disclosure is required by law or client agreement.
- Do not use AI tools to monitor, evaluate, or make decisions about employees without human review and documented criteria.


SECTION 4: DATA AND PRIVACY RULES

- Only use AI tools that have been approved by [designated role]. The current approved list is: [list tools here].
- Never paste client data, financial records, or personal information into AI tools that train on user input.
- When in doubt, anonymize data before using it with any AI tool.


SECTION 5: HUMAN REVIEW REQUIREMENTS

- All AI-generated content that will be shared externally (client deliverables, marketing, public communications) must be reviewed and approved by a human before publication.
- AI-generated code must be tested and reviewed before deployment.
- AI-generated analysis must be verified against source data before being used in decisions.


SECTION 6: CLIENT DISCLOSURE

- If a client asks whether AI was used in their work, answer honestly.
- For specific client contracts that prohibit or restrict AI use, follow the contract terms.
- Do not claim AI-assisted work is entirely human-created.


SECTION 7: BIAS AND FAIRNESS CHECKS

- When using AI for hiring, performance evaluation, or customer segmentation, review outputs for bias.
- If an AI tool produces results that seem discriminatory or biased, stop using it for that purpose and document the issue.


SECTION 8: TOOL APPROVAL PROCESS

- New AI tools must be submitted to [designated role] for approval before use.
- The approval considers: data handling practices, security certifications, training data policies, and business value.
- The approved tools list is reviewed quarterly.


SECTION 9: INCIDENT RESPONSE

- If an AI tool is used incorrectly or causes an issue (data leak, biased output, incorrect information shared with a client), report it to [designated role] within 24 hours.
- Document what happened, what was affected, and what corrective action was taken.
- Notify affected clients if their data was involved.


SECTION 10: REVIEW CADENCE

- This policy is reviewed quarterly and updated as AI tools and regulations evolve.
- All team members are responsible for following this policy.
- Questions can be directed to [designated role].

Copy the text above and replace the bracketed placeholders with your own details.