icon picker
AI Policies & Guidelines

AI Overview/Governance

Robert Hiett – Local Government AI Governance

This video discusses the integration of Artificial Intelligence (AI) in the public sector, emphasizing its potential to improve public service delivery by making operations more efficient and policymaking data-driven. It advocates for a gradual implementation of AI, starting with pilot programs to test and refine approaches, ensuring comprehensive training and adoption within the organization.

Example Policies and Guidelines

Sample Artificial Intelligence Policy
Union County | Artificial Intelligence Guidance Document

DOs

Use AI tools responsibly: You may use ChatGPT, Copilot, or similar tools as a starting point to brainstorm, outline, or draft ideas.
Review all outputs carefully: Always verify information for accuracy, bias, and copyright issues before using it in county work.
Create your own final copy: AI-generated content must be revised and rewritten — it cannot be the final version of a county document.
Protect county data: Only input public, non-sensitive information.
Ask IT or Legal if unsure: When in doubt about data sensitivity, copyright, or accuracy, contact the Director or Assistant Director of IT or Legal.
Stay informed: The County will continually review access to AI tools based on security and benefit — follow updated guidance.

🚫 DON’Ts

Do not input confidential data: Never share personally identifiable information (PII), public health information (PHI), or proprietary data (like employee info, software code, or internal reports).
Examples of banned inputs include:
Social Security numbers, driver’s licenses, credit cards, PINs, passwords, account numbers, biometric data, or parent’s maiden name.
Do not treat AI responses as fact: AI-generated information can be inaccurate or misleading — it must be fact-checked.
Do not upload or paste county documents without review: Make sure no sensitive data appears anywhere in a file before sharing it with an AI tool.
Do not rely on AI for final or official communications: All county work products must be reviewed and finalized by a human.
Do not ignore copyright risk: AI might reproduce copyrighted or proprietary material. Always confirm that outputs are original or public domain.
Do not assume privacy: Anything entered into AI tools is treated as public and may be stored or used by the company for model training.
Do not click on AI-crafted emails blindly: AI is now being used by cybercriminals to create realistic phishing messages. Stay alert.

🧭 Bottom Line

Use AI as a helper, not a decision-maker. Never input private, personal, or internal data. Always verify and humanize your final work before sharing or publishing.
Would you like me to turn this into a short one-page reference sheet (like a “cheat sheet” handout) for your

Email to add yours to this list.

Want to print your doc?
This is not the way.
Try clicking the ⋯ next to your doc name or using a keyboard shortcut (
CtrlP
) instead.