Overview

Lecture URL: https://skill.re/learn/manager/creating-team-ai-usage-guidelines.php

AI FOR MANAGERS CERTIFICATION

Responsible Independent Use (Level 3) | Responsible Independent Use

LECTURE: Creating Team AI Usage Guidelines

Lesson 3.5.5 | Estimated Duration: ~22 minutes

Welcome to the AI for Managers certification program. I am your instructor, and today we are covering one of the most important lessons in the Responsible Independent Use module: Creating Team AI Usage Guidelines.

This is Lesson 3.5.5 in Level 3, the AI Leadership track. Whether you are managing teams with established AI adoption, responsible for organizational standards, or providing guidance to multiple teams, the material in this session is designed to meet you where you are.

Today we focus on codifying your team's AI practices into living guidelines that serve as a reference, training tool, and governance document.

The goal is not to create bureaucratic red tape. The goal is to create practical guidance that helps your team use AI responsibly and consistently.

Before we begin, I encourage you to think about what guidelines your team needs to operate responsibly with AI. By the end of this session, you will know how to create them.

Let us get started.

Lesson 3.5.5: Creating Team AI Usage Guidelines

Purpose

As AI use becomes widespread in your team, guidelines ensure consistency, reduce risk, and codify learning. Guidelines are how you scale responsibility.

This lesson teaches you how to develop and maintain living guidelines for your team.

Why This Matters for Managers

A common mistake is assuming people will remember conversations and make consistent decisions. Without written guidelines, decisions become inconsistent and risk grows.

The stakes include:

• Consistency (everyone uses AI the same way)
  - Risk management (guidelines prevent misuse)
  - Onboarding (new people learn standards quickly)
  - Accountability (guidelines make expectations clear)
  - Continuous improvement (guidelines evolve with learning)

Your job is to create and maintain guidelines that serve your team.

Creating Your Guidelines Framework

Step 1: Assess Your Team's Current State

Before writing guidelines, understand:

• Which AI tools does your team use?
  - Which processes use AI?
  - What good practices already exist (and which are inconsistent)?
  - What risks or problems have emerged?
  - What does the team say they need guidance on?

Do not assume you know. Ask your team: "What guidance would help you use AI better?"

Step 2: Choose Your Structure

Guidelines typically follow this structure:

Section 1: Purpose and Principles

"These guidelines help us use AI responsibly, effectively, and consistently. We believe AI should be used to enhance human capability, not replace human judgment. We prioritize transparency, accuracy, and ethical use."

Section 2: Approved Tools

"These are the AI tools we have approved for use. Using other tools requires approval."

• ChatGPT Pro (for writing, analysis, brainstorming)
  - GitHub Copilot (for code assistance, within security guidelines)

Include: Tool name, approved use cases, security/compliance status

Section 3: Use Case Guidance

For each major use case (email drafting, code review, customer analysis, etc.):

• What is this use case?
  - When should we use AI? (When is it appropriate?)
  - When should we NOT use AI? (What are limits?)
  - How do we verify output? (Review checklist)
  - Who reviews? (Owner/role)
  - What data can we input? (Data privacy constraints)
  - Documentation: How do we document AI use?

Section 4: Data Protection

"What data is safe to input to AI tools? What is not?"

• Safe: Public information, internal non-sensitive work
  - Not safe: Customer personal data, financial information, passwords, proprietary algorithms

Section 5: Quality Standards

"How do we ensure quality of AI outputs?"

• Verification requirements for each use case
  - Who is responsible for verification?
  - Common error types to watch for
  - Escalation process if something is wrong

Section 6: Ethical Boundaries

"What uses of AI are off-limits?"

• No using AI to make high-stakes decisions about people
  - No using AI to analyze employee data without consent
  - No using AI to generate deceptive content
  - No using AI for discriminatory purposes
  - Ethical concerns: Escalate to manager

Section 7: Transparency and Disclosure

"When do we disclose AI use?"

• To customers: Always, if AI was used in deliverables

Section 8: Support and Escalation

"Who do I ask if I have questions?"

Section 9: Evolution and Feedback

"How do we improve these guidelines?"

Making Guidelines Actionable

Step 1: Use Case Depth

Generic guidelines are not helpful. Be specific to actual work.

Generic: "Review all AI outputs"

Actionable: "For customer emails, review for: accuracy of facts, tone appropriateness, and completeness. Verification should take 2-3 minutes per email. Sarah or Jordan review before sending."

Generic: "Use AI responsibly for writing"

Actionable: "For status updates to leadership, use AI to draft. Verify all facts against source docs. Review for tone--should be confident but not overstated. Usually 5 minutes of review time."

Step 2: Examples and Anti-Examples

For each major use case, provide examples:

Good use: "Sarah asked AI to brainstorm 10 ideas for improving customer onboarding. She reviewed the ideas, kept 3, and presented them to the team."

Bad use: "Alex asked AI to rank which team members are highest-risk performers. He used the rankings to decide who to put on performance improvement plans."

Anti-use: "Do not use AI to: Make hiring/firing decisions. Analyze employee emails without consent. Generate deepfakes or deceptive content."

Step 3: Integration into Workflows

Make guidelines visible where people actually work.

Don't: Create a 20-page document nobody reads

Do:

• One-page summary in Slack
  - Detailed guidelines in wiki (linked from Slack)
  - Checklist in project management tool before submitting AI-generated work
  - Examples in team meetings

Step 4: Clear Ownership

Someone owns the guidelines. Usually the manager, but can delegate to someone else.

Owner is responsible for:

• Maintaining accuracy
  - Updating based on new learning
  - Answering questions about guidelines
  - Periodic review (quarterly or semi-annually)

Step 5: Governance and Updates

Guidelines should evolve, but not change constantly.

Update process:

• Quarterly review: Is this still accurate? What have we learned?
  - Feedback loop: Team suggests changes throughout the quarter
  - Decision-making: Owner decides on updates with input from team
  - Communication: Major changes are announced; minor clarifications are noted
  - Version control: Track changes (version 1.0, 1.1, 1.2, etc.)

Example timeline:

• Months 1-3: Use Version 1.0, gather feedback
  - Month 3: Review and update to Version 1.1
  - Months 4-6: Use Version 1.1, gather feedback
  - Month 6: Review and update to Version 2.0
  - And so on...

Building Consensus on Guidelines

Guidelines are most effective when the team helps create them:

Step 1: Propose Initial Framework

"I want to create guidelines for our AI use. Here is a structure I am proposing."

Step 2: Gather Input

"What use cases should we cover? What rules do we need? What questions do people have?"

Step 3: Draft Together

Ideally, involve team members in drafting. At minimum, share draft for feedback.

"Here is a draft. What do you think? What is missing? What do you disagree with?"

Step 4: Refine

Incorporate feedback. Revise and re-share.

"Based on your feedback, here are changes I have made."

Step 5: Finalize and Launch

Step 6: Train and Support

Do not assume people read guidelines. Train on them.

"In our team meeting, we will walk through the guidelines. Questions?"

Teams that help create guidelines are more likely to follow them.

Common Guideline Elements Checklist

When creating guidelines, ensure you cover:

Scaling Guidelines Across Teams

If you are responsible for multiple teams:

Approach 1: Centralized Guidelines with Local Adaptation

• Create organizational guidelines
  - Allow teams to adapt for their context
  - Central review ensures consistency

Approach 2: Framework + Local Ownership

• Create a framework/template
  - Each team develops their own guidelines
  - Share learning across teams

Approach 3: Federated Governance

• Central principle-based guidelines (ethical boundaries, data protection)
  - Team-specific use case guidelines
  - Regular sync across teams

Choose the approach based on your organization's structure and culture.

ANTI-PATTERNS

Anti-Pattern 1: Guidelines Created Without Team Input

"I wrote the guidelines and am announcing them."

Why it fails: Team did not buy in. Adoption is slow.

Better: Collaborate on creation. Team helps shape guidelines.

Anti-Pattern 2: Guidelines That Are Too Vague

"Use AI responsibly. Apply good judgment."

Why it fails: Not actionable. People do not know what they are actually supposed to do.

Better: Specific guidance for specific situations.

Anti-Pattern 3: Guidelines That Are Too Restrictive

"Never use AI. It is too risky."

Why it fails: Team will ignore guidelines they see as unreasonable.

Better: Guidelines that allow responsible use while preventing misuse.

Anti-Pattern 4: Guidelines That Become Stale

"We created guidelines two years ago and have not touched them."

Why it fails: Guidelines no longer reflect reality. Team ignores them.

Better: Regular review and update. At least quarterly.

Anti-Pattern 5: Guidelines With No Support Structure

"Here are the guidelines. Figures it out."

Why it fails: People have questions. Without support, they either ignore guidelines or guess.

Better: Clear escalation process and support channels.

PRACTICE PROMPTS

1. Use Case Assessment: List five AI use cases on your team. For each, what guidance does the team need? (About when to use AI, how to verify, data protection, etc.)
2. Framework Design: Using the structure provided, sketch out the main sections of guidelines for your team.
3. Use Case Detail: Choose one major use case. Write detailed guidance: when to use AI, when not to, how to verify, who reviews, time to review.
4. Team Input: Draft a message to your team: "I want to create AI guidelines. What do you think we should cover?" Capture their feedback.
5. Evolution Plan: Design how you would update and evolve guidelines over time. What does quarterly review look like?

KEY TAKEAWAYS

1. Guidelines should be specific to your team's work, not generic rules.
2. Involve your team in creating guidelines. They will follow guidelines they helped create.
3. Make guidelines actionable: use cases, examples, clear ownership, escalation paths.
4. Integrate guidelines into workflow, not just a document people read once.
5. Evolve guidelines based on learning and feedback.

GLOSSARY

Guidelines: Written standards for how a team uses AI tools responsibly and consistently.

Use Case: A specific situation where AI is applied. Examples: email drafting, code review, data analysis.

Governance: The process of making and enforcing decisions about how to use AI.

Escalation: The process of raising a question or concern to someone with more authority or expertise.

Owner (or Process Owner): The person responsible for maintaining and updating guidelines.

[SYNTHESIS AND APPLICATION]

Let us step back and look at the bigger picture of what we have covered in this session on Creating Team AI Usage Guidelines.

Guidelines codify the learning your team has accumulated. They make responsible AI use the default, not the exception.

Here is what I want you to take away from this session:

First, ownership. Guidelines require clear ownership. Someone maintains them.

Second, evolution. Guidelines should evolve. What is true today might not be true in six months.

Third, engagement. Guidelines are most effective when the team helps create and maintain them.

[REFLECTION EXERCISE]

Before we close, I would like you to spend two minutes on this reflection:

If you were to create AI guidelines for your team today, what would be the three most important things to cover? What does your team most need guidance on?

Write down your answer. Those three things are where your guidelines should focus.

[CLOSING REMARKS]

In our final lesson, we will explore running hybrid meetings with AI support, bringing together many of the skills you have learned. You have created guidelines. Now we will discuss how to apply them in a specific context.

This has been Lesson 3.5.5: Creating Team AI Usage Guidelines, part of the Responsible Independent Use module in Level 3: AI Leadership of the AI for Managers certification.

Remember: guidelines are not constraints. They are enablers. Good guidelines help your team use AI confidently and consistently.

Thank you for your time, your attention, and your commitment to building a responsible AI culture in your team.

END OF TRANSCRIPT

AI for Managers Certification Program

Level 3: AI Leadership | Responsible Independent Use | Lesson 3.5.5

A SkillsClinic initiative by No Worker Left Behind and The Work Company.

Duration: ~22 minutes | Word Count: ~3412