A quick note: You’re receiving this email because we’ve connected recently about AI, leadership, or communications/marketing, and I thought you might find these field notes useful. I’ll send this about once a month. If it’s not helpful, you can unsubscribe anytime using the link at the bottom—no hard feelings.
THIS MONTH'S FIELD SIGNAL
Why AI Mandates Rarely Stick
I’ve been on a listening tour the last few months, talking with senior leaders across comms, marketing, tech, and professional services who are under pressure to “do something with AI” and make it stick—not just launch another tool.
In those conversations, there’s a pattern that keeps showing up. Most organizations start their AI journey the same way: with a mandate and a tool.
A mandate (“Everyone needs to start using AI.”)
A tool (Copilot, ChatGPT, an internal GPT, take your pick)
What tends to get overlooked is the hard, human work in between: aligning leaders, equipping managers, and creating the conditions where people feel safe enough to actually change how they work.
Leaders assume that once people understand what the tools can do and have access to them, adoption will naturally follow. But in real organizations, that’s rarely how change works—especially change that touches identity, job security, and professional judgment.
What I’m hearing instead sounds like this:
People know AI matters, but they’re unclear how it applies to their role, their team, or their future in the organization.
Managers want to support experimentation, but don’t know what’s allowed or expected.
Teams get training, try a few things, then quietly revert to old habits—especially if the AI doesn't work perfectly.
Six months later, executives are asking why adoption has stalled, and why the investment story isn’t matching the reality on the ground.
This isn’t a tooling problem. It’s a human one—and that’s where most of the cultural and reputational risk actually lives.
Most AI rollouts skip over the hardest (and most important) questions:
Do people actually want this change?
Do they trust how it’s being framed?
Do they feel safe experimenting?
And who is reinforcing the new behaviors once the novelty wears off?
AI adoption doesn’t stall because people are incapable. It fails because organizations underestimate how much belief, clarity, and reinforcement it takes for change to stick.
FROM THE FIELD
A Real-World Example
In recent conversations with two leaders at the same large organization—one in communications, one with a broader view across the business—I heard the same story from different angles: strong AI intent on paper, weak human conditions in practice.
From the communications side, the message was clear: AI was a priority, and teams were expected to be using it more. What was missing was clarity. Which tools were actually okay to use? What problems were they supposed to focus on? And how would anyone know if they were “doing it right”? People nodded along in meetings, then went back to their desks to quietly figure it out on their own. That silence looked like alignment from the top but felt like risk and ambiguity on the ground.
From elsewhere in the organization, the pattern looked different but related. After years of constant change, people had learned to be cautious. Speaking up or taking risks didn’t feel safe—even when leaders said they wanted innovation. The actual signals (who gets praised, who gets penalized, what gets resourced) rewarded predictability—not experimentation that might fail.
Put together, these perspectives explain why AI adoption stalled. The mandate was there. The tools were there. But the human conditions weren’t. Without clear intent, psychological safety, and ongoing reinforcement, people defaulted to the safest behavior available: do what you know, don’t draw attention, and wait for clearer signals. That’s a leadership design problem, not an employee motivation problem.
QUESTIONS FROM THE TRAIL
Ask Yourself
These prompts are meant to help you reflect, or spark a practical conversation with your team:
Where are we assuming enthusiasm for AI instead of actively creating it?
What fears or tradeoffs might people be carrying that we haven’t acknowledged out loud?
What signals are our leaders actually sending about AI—through priorities, metrics, and time, not just words?
After training ends, what are we doing to reinforce new behaviors—or are we assuming they’ll stick on their own?
How are we equipping managers to coach responsible AI use, not just encourage tool usage?
ONE SMALL NEXT STEP
Recognize This Pattern?
If this pattern sounds familiar, I’d love to hear where you’re seeing it show up—just hit reply. And if you want to explore it more deeply, I offer a free 45-minute consult for senior leaders who are responsible for AI outcomes but concerned about the human side.
I often get asked what new AI tools I've been using recently—especially for offloading the kind of busywork that keeps leaders and teams from the deeper human work of change.
My latest favorite is Comet, an AI-enabled web browser from Perplexity that can act as a personal assistant.
You can ask it for help with whatever you’re looking at in the browser:
"Scan these three vendor pages and give me a quick brief: strengths, risks, and what I should ask in our next meeting."
"Look at our last two earnings calls, this analyst note, and our AI press release—what questions are investors most likely to ask me next?"
Even better, you can give Comet an assignment—pretty much anything you could do yourself in a browser—and let it take over the work from there. I've recently had it:
Fix a technical problem on my web hosting platform that was over my head.
Copy info from LinkedIn, Hubspot and ChatGPT and consolidate it in one place to help me prep for meetings.
For leaders and their teams, this kind of delegated work lets you move from “I don’t have time to think about AI” to “I have help handling the noise so I can focus on judgment, decisions, and signaling what matters to this organization right now.”
Trailhead Communications, 7327 SW Barnes Rd #1014, Portland, OR
