Building a Learning Culture

Overview

A learning culture around AI doesn't require formal programs. It requires psychological safety and shared curiosity.

Psychological safety: People feel safe sharing failures. If you tried something with AI and it didn't work, can you talk about it without embarrassment? In a learning culture, yes.

Shared curiosity: People approach AI with experimentation mindset. "Let's see what works" rather than "I need to be right."

Openness to others' ideas: People listen to colleagues' discoveries and test them, rather than dismissing or ignoring.

These create an environment where collaborative learning happens naturally.

Content

Discoveries come in different forms. Here's how to share each effectively:

Discovery Type 1: "I Found a Better Prompt Structure"

You've discovered a way of prompting AI that generates much better output.

How to share:

• Describe the problem you were trying to solve
• Describe what you tried initially (and why it was ineffective)
• Describe the new approach
• Provide a specific example showing the difference

Example:

"I was asking AI to draft responses to complaints. I was getting corporate-sounding output that didn't match our tone. I tried changing my prompt from 'Draft a professional response' to 'Draft a response that sounds like [agent name] talking directly to this customer. Warm, personal, but professional.' Results were way better. Here's an example of the difference..."

Discovery Type 2: "I Found a Limitation"

You've discovered a situation where AI fails or produces unreliable output.

How to share:

• Describe the situation
• Describe what you expected AI to do
• Describe what actually happened
• Explain why you think it failed
• Provide a suggestion for how others should handle similar situations

Example:

"I tried using AI to predict whether a customer would escalate based on their message tone. AI was wrong a lot. It would predict escalation when the customer was just frustrated, or miss actual escalation indicators. I think it's because predicting behavior requires context AI doesn't have. I'd suggest we not rely on AI for this, but use it to flag emotions in messages instead."

Discovery Type 3: "I Found a Use That Works Really Well"

You've discovered a task where AI is genuinely helpful and reliable.

How to share:

• Describe the task
• Explain why this task is good for AI
• Describe how you use it (high-level process)
• Describe the benefit
• Offer to help others try it

Example:

"I've started using AI to help me draft follow-up questions when a customer message is vague. Instead of just asking for clarification generally, I ask AI to generate 3-5 specific questions that would help clarify the situation. AI is good at this because it's not requiring judgment, just generating options. Saves me time and customers appreciate more thoughtful clarification questions."

How to Share Discoveries

You've discovered something risky about how AI is being used.

How to share:

• Describe the situation
• Explain why it's risky
• Provide a specific example if possible
• Suggest how to handle it more safely

Example:

"I noticed someone pasting full customer conversations with account numbers into AI to ask for response suggestions. The account number shouldn't be shared with external AI systems. I mentioned it to them. We might want to remind the team about data sensitivity when using external AI."

Informal Learning Networks

A learning network isn't a formal program. It's colleagues talking about what they're learning.

Ways to create informal networks:

One-on-one conversations: "Hey, I found something interesting with how I'm using AI. Want to hear about it?"

Team huddles: Dedicate 5 minutes in team standup to "AI discovery of the week."

Shared notes: A shared document or channel where people post discoveries.

Lunch-and-learn style: Someone shares a discovery with small group. 15 minutes, informal, question and answer.

Peer observation: Watch a colleague use AI, ask questions, learn from how they approach it.

Experiment cycles: "This week, we're all trying to use AI for summarization. Let's compare notes Friday."

These are low-pressure ways to share and learn together.

Giving and Receiving Feedback on AI Use

Overview

As collaborative learning develops, colleagues will give each other feedback on how you're using AI.

How to give feedback helpfully:

Specific not general: "I noticed when you drafted that response, you didn't verify the policy against our knowledge base. Our policy has changed since AI's training data." Not: "You should verify things more."

Curious not judgmental: "I'm curious why you decided to use AI for that escalation decision. What was your thinking?" Not: "That's risky. AI shouldn't make that call."

Offer help not criticism: "This is something I struggled with too. Here's what helped me..." Not: "You're doing it wrong."

Focus on the behavior not the person: "That prompt didn't give AI enough context" not "You're not good at prompting."

How to receive feedback:

Listen without defending: Someone gives feedback. Your impulse is to explain why you did it that way. Resist. Listen first.

Ask clarifying questions: "Can you give me an example of what you mean?"

Thank them: Even if you disagree, thank them for caring about the quality of your work.

Reflect before deciding: You don't need to agree immediately. Reflect on what they said. Decide whether it changes your approach.

Anti-Pattern 1: Not Sharing Failures

People only share successes. Failures are hidden. As a result, the whole team learns more slowly. People make the same mistakes independently.

Better approach: Create safety around failure. Share what didn't work so others don't have to learn it the hard way.

Anti-Pattern 2: Sharing Without Context

Someone says "AI is great at summarization" without explaining when or how. Others try it in situations where it doesn't work as well. They get discouraged.

Better approach: Share with enough context that others understand when and how to apply what you learned.

Anti-Pattern 3: Not Acting on Shared Discoveries

Someone shares something valuable. Others don't try it. The discovery doesn't spread.

Better approach: When someone shares, commit to trying it for a week. Report back what you found.

Anti-Pattern 4: Hoarding Good Finds

You discover something that makes your work dramatically more efficient. You keep it secret to stay ahead of colleagues.

Problem: This prevents team improvement and undermines collaborative culture.

Better approach: Share your wins. The whole team improving makes everyone's work better.

Anti-Patterns: Collaborative Learning Failures

You share with your friends but not with colleagues you're neutral about. Information stays siloed.

Better approach: Create channels where discoveries reach the whole team.

Building Collaborative Habits

Start small:

Week 1: Share one discovery. Just one. Describe a situation where AI worked well or failed. See how people respond.

Week 2: Listen when colleagues share. Ask questions. Show genuine interest.

Week 3: Share another discovery. Ask a colleague about their experience with AI.

Week 4+: Continue the conversation. Slowly, a learning culture develops.

You don't need buy-in from leaders or formal programs. You need a few people who are curious and willing to share.

Practice Prompts

Prompt 1: Identify Your Discoveries

What have you learned about AI that would be valuable for colleagues to know? What works well? What are the limitations? What surprised you?

Prompt 2: Plan Your First Share

Choose one discovery. How would you describe it to colleagues? What context is important? What's the most helpful way to share this?

Prompt 3: Create a Learning Opportunity

How could you create a low-pressure space for your team to share AI discoveries? What would work for your team culture?

Prompt 4: Receive Feedback Gracefully

Imagine a colleague gives you feedback on how you're using AI. How would you respond? How would you reflect on it?

Key Takeaways

One. Collaborative learning accelerates team capability beyond what individual learning achieves.

Two. Share discoveries with context: the situation, what you tried, what worked, why.

Three. Create psychological safety so people share failures, not just successes.

Four. Informal learning networks work better than formal programs for cultural change.

Five. Give feedback with curiosity and helpfulness. Receive feedback with openness.

Six. Create channels where information reaches the whole team, not just certain people.

Seven. Collaborative learning starts with one or two people sharing. It grows organically from there.

Glossary

Psychological Safety: Feeling safe to take interpersonal risks without fear of negative consequences.

Learning Culture: Environment where people approach challenges with curiosity and share learning.

Collaborative Learning: Learning as a group rather than individually.

Informal Network: Group connected by shared interest, not formal structure.

Context: Information necessary to understand how a discovery applies.

Feedback: Information about performance or behavior offered with intention to help improve.

Reflection Exercise

Reflect on this: What discovery about AI have you made that would genuinely help colleagues? What's holding you back from sharing it? What would help you share?

Starting Small

If your organization doesn't have a collaborative learning culture yet, you can start building one:

Week 1: Share one discovery with one colleague. Start small.

Week 2: Mention something you learned in a team meeting. See how people respond.

Week 3: Ask a colleague about their experience with AI. Listen genuinely.

Week 4: Create a simple format for sharing (email thread, shared note, brief conversation). Make it easy.

Slowly, if people see that sharing is safe and useful, the culture grows. You don't need organizational buy-in to start. You need people willing to share.

Remote and Distributed Teams

If your team is distributed, collaborative learning still works:

Asynchronous sharing: A Slack channel, shared document, or email thread where people post discoveries.

Virtual discussions: Occasional video calls where people share and discuss.

Recorded learnings: Someone discovers something, records a 5-minute video explaining it. Team watches and learns.

Pair problem-solving: Two people tackle an AI problem together on a video call. Both learn from the approach.

Distribution doesn't prevent collaboration. It just requires more intentional setup.

Overcoming Barriers

Some people hesitate to share for understandable reasons:

Fear of seeming incompetent: If I share a failure, people will judge me.

Solution: Create psychological safety explicitly. Model sharing your own failures.

Competitive environment: If I share my wins, colleagues get ahead.

Solution: Frame it as team improvement, not individual competition.

Time pressure: I don't have time to share what I learn.

Solution: Make sharing quick and easy. 2-minute format beats 30-minute formal presentations.

Different experience levels: Some people know way more than others.

Solution: Frame it as learning from wherever you are. Beginners and experts both have valuable discoveries.

Address these barriers and more people contribute.

Closing Remarks

The teams that thrive with AI are those where people learn together, share discoveries, and help each other improve. Not in competition for who's best at AI, but in collaboration toward team excellence.

You don't need permission to start building this culture. You need curiosity and a willingness to share what you learn. Often, one person starting creates momentum that spreads.

In our next lesson, we'll explore measuring your own progress in AI skills.

AI for Customer Support Certification

Level 2: Assisted Use | Team Knowledge Sharing | Lesson 2.5.3

A SkillsClinic initiative.

Duration: ~25 minutes | Word Count: ~3,200

Key Takeaways

One. Collaborative learning accelerates team capability beyond what individual learning achieves.

Two. Share discoveries with context: the situation, what you tried, what worked, why.

Three. Create psychological safety so people share failures, not just successes.

Four. Informal learning networks work better than formal programs for cultural change.

Five. Give feedback with curiosity and helpfulness. Receive feedback with openness.

Six. Create channels where information reaches the whole team, not just certain people.

Seven. Collaborative learning starts with one or two people sharing. It grows organically from there.

Glossary

Psychological Safety: Feeling safe to take interpersonal risks without fear of negative consequences.

Learning Culture: Environment where people approach challenges with curiosity and share learning.

Collaborative Learning: Learning as a group rather than individually.

Informal Network: Group connected by shared interest, not formal structure.

Context: Information necessary to understand how a discovery applies.

Feedback: Information about performance or behavior offered with intention to help improve.

Reflection Exercise

Reflect on this: What discovery about AI have you made that would genuinely help colleagues? What's holding you back from sharing it? What would help you share?

Starting Small

If your organization doesn't have a collaborative learning culture yet, you can start building one:

Week 1: Share one discovery with one colleague. Start small.

Week 2: Mention something you learned in a team meeting. See how people respond.

Week 3: Ask a colleague about their experience with AI. Listen genuinely.

Week 4: Create a simple format for sharing (email thread, shared note, brief conversation). Make it easy.

Slowly, if people see that sharing is safe and useful, the culture grows. You don't need organizational buy-in to start. You need people willing to share.

Remote and Distributed Teams

If your team is distributed, collaborative learning still works:

Asynchronous sharing: A Slack channel, shared document, or email thread where people post discoveries.

Virtual discussions: Occasional video calls where people share and discuss.

Recorded learnings: Someone discovers something, records a 5-minute video explaining it. Team watches and learns.

Pair problem-solving: Two people tackle an AI problem together on a video call. Both learn from the approach.

Distribution doesn't prevent collaboration. It just requires more intentional setup.

Overcoming Barriers

Some people hesitate to share for understandable reasons:

Fear of seeming incompetent: If I share a failure, people will judge me.

Solution: Create psychological safety explicitly. Model sharing your own failures.

Competitive environment: If I share my wins, colleagues get ahead.

Solution: Frame it as team improvement, not individual competition.

Time pressure: I don't have time to share what I learn.

Solution: Make sharing quick and easy. 2-minute format beats 30-minute formal presentations.

Different experience levels: Some people know way more than others.

Solution: Frame it as learning from wherever you are. Beginners and experts both have valuable discoveries.

Address these barriers and more people contribute.

Closing Remarks

The teams that thrive with AI are those where people learn together, share discoveries, and help each other improve. Not in competition for who's best at AI, but in collaboration toward team excellence.

You don't need permission to start building this culture. You need curiosity and a willingness to share what you learn. Often, one person starting creates momentum that spreads.

In our next lesson, we'll explore measuring your own progress in AI skills.

AI for Customer Support Certification

Level 2: Assisted Use | Team Knowledge Sharing | Lesson 2.5.3

A SkillsClinic initiative.

Duration: ~25 minutes | Word Count: ~3,200