There's a conversation I have at least once a week. A business leader tells me their team "doesn't get AI." They say it with frustration, like the team is the problem. But when I dig in, the real issue is almost always the same, nobody ever gave the team a reason to understand it.
People don't resist AI because they're stupid or stubborn. They resist it because it was introduced as a mandate, not a conversation. "We're implementing AI" is a very different message than "Here's a problem we all deal with, and here's how this tool might help. What do you think?"
AI fluency starts with that shift. It's not a training program. It's a culture change.
Three Levels of Fluency
Not everyone in your organization needs the same level of understanding. That's important. Trying to teach everyone everything is how you waste everyone's time. Think of it in three levels.
Awareness (Everyone)
What AI is, what it isn't, and how it relates to their work. This is the "no surprises" level. When the company announces an AI initiative, people at this level understand what that means at a basic level. They aren't scared or confused. They have enough context to ask reasonable questions.
Takes: A few sessions over 2-3 weeks. Not a course. Conversations, demos, Q&A.
Competence (Users & Managers)
How to use AI tools effectively and evaluate their outputs. This is the "can I trust this result?" level. People here know how to prompt AI systems, how to verify outputs, when to trust the tool, and when to double-check. They understand the limitations relevant to their specific work.
Takes: 4-6 weeks of hands-on work with real tools on real tasks, plus ongoing practice.
Strategy (Leadership & Champions)
How to evaluate AI opportunities, manage risk, and make investment decisions. This is the "should we do this?" level. People here can run a CLEAR evaluation, assess vendors, calculate real ROI, and communicate AI strategy to stakeholders.
Takes: Ongoing engagement. This isn't a destination. It's a practice.
Five Mistakes That Kill AI Fluency Programs
1. Starting with technology instead of problems
"Here's what a transformer model does" means nothing to your operations team. "Here's how we could cut the time you spend on monthly reporting by 60%" means everything. Start with their pain, not your tech stack.
2. Making it a one-time event
A lunch-and-learn is not a fluency program. It's a box to check. Real understanding builds over time through repeated exposure, hands-on use, and ongoing conversation. If your "AI training" fits in a single calendar block, it's not training.
3. Excluding skeptics
The people who are most skeptical about AI are often the ones who understand the business processes best. They're not resisting change for fun. They're worried about real things, job security, reliability, quality. Include them early. Listen to their concerns. They'll become your best advocates or your most valuable quality checkers. Either outcome is a win.
4. Not giving people safe space to experiment
People need to use AI on low-stakes tasks before anyone trusts them with high-stakes ones. If the first time someone uses an AI tool is on a production customer interaction, you've set them up to fail. Create sandbox environments. Let people play, make mistakes, and learn without consequences.
5. Treating fluency as IT's job
AI fluency is not a technical initiative. It's an organizational one. If it's owned by IT alone, it will be perceived as a tech project that the rest of the company needs to "deal with." Ownership needs to come from leadership, with IT as an enabler, not the driver.
A Practical 90-Day Approach
Here's what a realistic AI fluency initiative looks like. Not a theoretical ideal. Something you can actually do.
Days 1-30: Foundation
- Identify 3-5 real business problems that AI might address
- Run awareness sessions for the broader team (what AI is, what it means for them)
- Identify 2-3 "AI champions" in different departments (not just IT)
- Give everyone access to a basic AI tool (ChatGPT, Claude, etc.) with usage guidelines
Days 31-60: Exploration
- Champions run hands-on experiments with AI on the identified problems
- Track what works, what doesn't, and what surprised people
- Leadership attends strategy sessions (CLEAR framework, vendor evaluation, ROI)
- Share findings openly. Both successes and failures. Especially failures.
Days 61-90: Decision
- Run CLEAR on the most promising opportunity
- Develop a realistic proposal with real numbers (using the cost framework from our cost article)
- Make a go/prepare/wait decision based on evidence, not enthusiasm
- Plan Phase 2 (whether that's implementation, more preparation, or revisiting in 6 months)
The Real Goal
The goal of AI fluency isn't to make everyone an AI expert. It's to create an organization where AI decisions are informed, where concerns are heard, and where adoption happens because people understand the value, not because they were told to.
That kind of organization doesn't just adopt AI better. It adopts every technology better. Because the real skill you're building isn't "AI knowledge." It's organizational intelligence.
"The smartest organizations aren't the ones with the most AI tools. They're the ones where everyone can explain why they chose the tools they have."
- Daryl Lantz, MindXpansion