The Confidence Gap: Why Leaders Who Build Capability Outperform Those Who Wait

I watched a CEO freeze mid-sentence during a board call last week. The question was simple: “How is AI changing your operating model?” The silence stretched until someone mercifully moved to the next agenda item.

That moment crystallized something I’ve been observing across dozens of leadership conversations: there’s a widening gap between leaders who build capability and those who wait for certainty. And in 2026, that gap isn’t just about competitive advantage—it’s about survival.

The “Wait and See” Trap

Here’s the uncomfortable truth: many leadership teams still think they can “wait and see” while their operating models change underneath them. I see it everywhere—from family businesses in the Mittelstand to PE-backed growth companies.

The logic feels reasonable: “Let’s see how this AI thing shakes out. Let’s wait for clearer regulations. Let’s see what our competitors do first.”

But confidence doesn’t come from having all the answers. Confidence comes from competence, and competence comes from doing.

In my experience, the firms winning right now aren’t the ones with the best slogans or the biggest AI budgets. They’re the ones building capability systematically:

• Clear AI use cases tied to measurable value
• Training that creates reps, not slide decks
• Governance frameworks that let teams move fast without breaking trust
• Leaders who actually learn the tools instead of delegating “AI” to IT

The Sponsorship Principle

This capability-building mindset connects to something I’ve been thinking about since International Women’s Day: the difference between mentorship and sponsorship.

Mentors advise. Sponsors create opportunities when you’re not in the room.

The same principle applies to organizational capability. Leaders who truly transform their businesses don’t just advise their teams to “be more innovative.” They sponsor capability by:

• Putting people into visible, high-stakes AI projects with real decision rights
• Backing them publicly when experiments don’t go perfectly
• Creating space for learning that isn’t disguised as performance theater

I think of it as “Trust Infrastructure”—you build the systems that allow people to take intelligent risks and learn fast.

The Underpromise Framework

Which brings me to what I call the Underpromise Accelerator—a simple pattern I’ve seen work across industries:

Step 1: Underpromise, Overdeliver (Consistently)
Start with pilot projects that seem almost too small. Hit every milestone. Build trust through reliability before you ask for bigger bets.

Step 2: Make Impact Visible (Without Being Loud)
Communicate outcomes, not effort. “We reduced manual processing time by 40% in accounts payable” hits differently than “We’re really focusing on AI integration.”

Step 3: Stay Human
Technology amplifies culture, it doesn’t create it. The leaders I respect most treat AI adoption like any other change management challenge—with patience, clarity, and genuine concern for their people.

This isn’t about moving fast and breaking things. It’s about building the muscle to adapt quickly while maintaining trust.

The Real Infrastructure Play

Here’s my contrarian take: AI isn’t a technology play—it’s a productivity infrastructure play. Just like you wouldn’t run a modern business without internet or email, you won’t run a competitive business in 2028 without embedded AI capability.

The difference is that internet adoption happened gradually over decades. AI capability is compressing into months and years.

Early in my career at Bertelsmann, I had a leader—Ralf Bierfischer—who taught me something that still guides how I think about organizational change: high standards and human respect aren’t opposites. The best leaders set the bar high precisely because they believe in their people’s ability to reach it.

That’s the mindset we need for AI adoption. Not “this is too complex for our team” but “how do we build the capability our team deserves?”

Your Next Week Action Plan

If you’re ready to close the confidence gap, here are five things you can do starting Monday:

1. Pick One Manual Process: Identify your team’s most repetitive, frustrating task. Don’t automate it yet—just document it clearly.

2. Schedule 30 Minutes with AI: Use ChatGPT, Claude, or similar tools for one actual work task (not a demo). See what happens.

3. Ask Your Team: “What would make you 20% more effective at your core job?” Listen for patterns.

4. Review Your Meeting Rhythms: Are you spending more time talking about AI than experimenting with it?

5. Find Your Sponsor: Identify one person in your organization who needs visible backing on a capability-building project. Give it to them.

The leaders who thrive in the next 24 months won’t be the ones who had perfect information. They’ll be the ones who built capability while others waited for certainty.

Question for reflection: What’s one area where your organization is “waiting and seeing” when you could be “learning and building” instead?