From Individual to Team Fluency
Why individual AI skill doesn't add up to team fluency, what 'shadow AI' costs, and the leader's role in turning scattered experiments into shared practice.
Ten Fluent People Aren’t a Fluent Team
Picture a team where everyone uses AI, and uses it well, each in their own way. One person drafts client emails with a personal chatbot. Another summarizes meeting recordings with a browser extension. A third built a clever spreadsheet automation nobody else knows about.
Individually, they’re productive. As a team, there are problems:
- Nobody knows what data has gone where. Client details may be sitting in personal accounts with unknown data terms.
- Quality is inconsistent. One person checks AI output carefully; another doesn’t check at all. Customers can’t tell the difference until something goes wrong.
- Good practices don’t spread. The best prompt or workflow stays in one person’s head.
- Accountability is fuzzy. When AI-assisted work has an error, nobody is sure what the review process was supposed to be.
This is often called shadow AI: AI use that happens outside any shared visibility or rules. It usually isn’t anyone being reckless. It’s what happens when people find useful tools and nobody has set shared expectations.
The 4Ds at Team Level
The 4D AI Fluency Framework applies to teams as well as individuals, but the questions change:
| Competency | Individual question | Team question |
|---|---|---|
| Delegation | Should I use AI for this task? | Where does AI fit in our workflows, and where do we require human judgment? |
| Description | How do I explain what I need? | What shared context, instructions, and standards do we give AI? |
| Discernment | Is this output good? | Who reviews what, to what standard, and how do we learn from mistakes? |
| Diligence | Am I using AI responsibly? | What are our rules for tools, data, disclosure, and accountability? |
- 1 Scattered
Individual use, no visibility
- 2 Visible
We know who uses what, for what
- 3 Shared
Common tools, instructions, and review
- 4 Fluent
Deliberate, improving, accountable
The Leader’s Role
You don’t need to be the team’s best AI user. Your job is different:
- Make it safe to talk about. People hide AI use when they fear judgment or think it’s against the rules. You can’t improve what you can’t see.
- Set direction. Decide, with the team, where AI belongs and where it doesn’t.
- Create shared resources. Approved tools, shared instructions, review standards.
- Model it. Use AI openly and well, including showing when you checked it, corrected it, or decided not to use it.
- Own the outcomes. AI-assisted work from your team is your team’s work.
Common mistake: either banning AI outright (“too risky”) or pushing everyone to “use AI for everything” to hit a productivity target.
A ban drives use into the shadows, where risk is highest. Hype pushes AI into tasks where it degrades quality or bypasses judgment. Both skip the real work: deciding, together, where AI genuinely helps.
- Send your team a short, anonymous survey (or hold an open conversation, if trust is high). Ask:
- Which AI tools do you use for work, if any?
- What do you use them for?
- What kind of information do you put into them?
- Where has AI helped most? Where has it caused problems?
- What would make you more confident using AI at work?
- Make clear that the goal is to learn, not to punish.
- Summarize the results: tools, uses, data types, wins, and concerns.
- Mark where your team sits on the path above: Scattered, Visible, Shared, or Fluent.
1. What is 'shadow AI'?
Shadow AI is unseen, unmanaged use. It's usually the natural result of useful tools without shared expectations.
2. Why don't ten individually skilled AI users automatically make a fluent team?
Team fluency needs shared delegation decisions, instructions, review standards, and rules.
3. What's the first thing a leader should do to move from scattered to shared AI use?
You can't improve what you can't see. Visibility comes from trust.
About the Framework
The 4D AI Fluency Framework was created by Rick Dakan and Joseph Feller with Anthropic, and is released under CC BY-NC-SA 4.0. This course’s lessons, examples, and exercises are original. For the individual-level framework, see AI Fluency: Foundations.
Up Next
With a picture of how your team uses AI today, the next step is deciding where it should fit: mapping your team’s actual work.