Pillar 2: Human-Centered AI Transformation

Human-Centered AI Transformation

I know it seems counterintuitive - but I promise it's possible.

Employees don't resist AI. They resist bad communication about AI.

In the past two years, I've operationalized AI adoption communications twice - at two different organizations, with two very different workforces.

Both times, we measured it. Both times, it worked.

I'm sharing what I learned because the narrative in most boardrooms right now is dangerously wrong - and it's costing organizations the very buy-in they're desperate for.

The wrong diagnosis:

When AI rollouts stall, leadership typically reaches one of three conclusions:

• "Employees are afraid of losing their jobs."

• "People just resist change."

• "We need more training."

  
  

These aren't wrong, exactly. But they're downstream of the real problem.

The real problem is this: only 14% of employees feel their organization is ahead on AI. That's not a training gap. That's a trust gap. And you cannot close a trust gap with a training program.

99% of employees in corporate america in 2026 are afraid they are being forced to train AI so the company can replace them with the AI that they trained. ( I made up that statistic but I swear a LinkedIn poll would prove me correct) (also - a trust gap)

Simon Sinek's research into what makes organizations inspire rather than simply inform is relevant here in a way that most AI rollout teams miss entirely. In Start With Why, Sinek documents how the organizations that actually move people - that generate loyalty, engagement, and real behavior change - communicate from the inside out.

They start with why, not what.

Most AI rollouts do the opposite: they lead with the tool, the capability, the timeline, the mandate. They skip the part where employees understand what it means for them, why leadership made this decision, and whether the organization is being honest about the uncertainty involved. Employees don't buy into the what. They buy into the why — and only if they believe the people telling them have been straight with them before.

What I saw in the real world:

The employees I worked with weren't afraid of AI in the abstract. They were afraid of specific things that no one had addressed:

"Will I be judged for using it wrong?" "Is this replacing my team or augmenting it - and is anyone actually going to tell me the truth?" "If I raise concerns, will I be seen as the problem?" "The last three 'transformations' didn't stick. Why is this one different?"

These are not irrational fears. They are entirely reasonable responses to organizations that have historically communicated change in ways that prioritized narrative control over genuine transparency.

The employees who disengaged from AI adoption weren't intentionally being difficult. They were 25+ year industry veterans whose entire position existed because of fraud and they were taught to be distrustful of outside data and distrustful of computers doing what only they could do. They were people whose trust had been eroded by past communication that turned out to be spin. And their nervous systems had gotten very good at detecting the pattern.

What we did differently:

We didn't launch with a hype campaign. We launched with honest conversation.

We named the uncertainty directly. We acknowledged what we didn't know. We created channels - real ones, not performative ones - for employees to raise questions without those questions being filtered through a manager who might feel threatened by them. We provided end users sandboxes to test in and actually gathered their feedback.

We designed the communications around three behavioral principles:

1. Psychological safety first. Before we asked anyone to adopt anything, we established that it was safe to be uncertain. That it was safe to try and fail. That questions wouldn't be penalized.

2. Relevance over volume. Instead of a company-wide AI rollout deck, we built role-specific communications. What does AI mean for your job, your workflow, your day? The more specific the message, the more it landed. We also included the long term strategic plan so they knew we weren't trying to replace them.

3. Social proof from inside. We identified early adopters >>> not management champions, but peers << and let them tell the story. Behavioral science is unambiguous on this: people change their behavior when people like them change theirs. Top-down mandates don't move culture. Horizontal trust does.

The results:

Adoption rates climbed. Skepticism dropped. not because we suppressed it, but because we gave it a legitimate place to live. And the employees who had been most resistant became, in some cases, the most vocal advocates because someone had finally talked to them like adults.

What this means for your organization:

If your AI adoption is stalling, the answer probably isn't a better deck or a flashier launch event.

It's a harder, more honest conversation.

One that acknowledges the gap between what leadership knows and what employees experience. One that treats resistance as data rather than a PR problem. One that's designed around human behavior, not around the initiative's timeline.

The organizations that win on AI adoption won't be the ones with the best technology. They'll be the ones whose employees actually use it, because they understood that adoption is a people problem before it's ever a technology problem.

There will always be new technology - the best leaders know that people is what matter most.

Sources referenced:

• Sinek, S. (2009). Start With Why: How Great Leaders Inspire Everyone to Take Action. Portfolio/Penguin.

• Brown, B. (2018). Dare to Lead: Brave Work, Tough Conversations, Whole Hearts. Random House.

• Edmondson, A.C. (1999). "Psychological Safety and Learning Behavior in Work Teams." Administrative Science Quarterly.

• Edmondson, A.C., Bahadurzada, H. & Kerrissey, M. (2024). "Psychological Safety as an Enduring Resource Amid Constraints." International Journal of Public Health.