For the past few years, the dominant narrative about AI has focused on what it will replace and not nearly enough on what it will make possible.

AI will eliminate some roles and change many more. Let’s just say that out loud. Three roles will become one. Ten roles will become three. As they always have, CEOs will need to make hard calls about where to invest and where to pull back. We’ve made some of those calls at BTS, and AI will undoubtedly require us to make more.

But if efficiency is the beginning and end of our conversation with employees, we shouldn’t be surprised when they’re reluctant to experiment with AI, much less push the boundaries of what our companies are capable of. We need to think about AI as a strategic opportunity to serve customers better, to take market share, to increase growth, in addition to expanding margins. It’s far more important that we use AI to think creatively and exponentially—restructuring workflows, reducing daily drudgery, developing new business opportunities and building new products and services—rather than just cutting costs.

I’ll be the first to admit I don’t have all the answers. No CEO does. But I do know this: Our top AI priority is to teach our people the power of AI, create an innovation mindset and challenge them to improve their operations—higher quality, faster cycle times and lower cost. Employees will lead the way to of figuring this out—they know the business better than anyone. The future of business will be shaped by how well we mobilize our teams to create breakthoughs.

1. Start with discovery

One of the biggest mistakes I see organizations making is asking employees to hit a cost-reduction or productivity target before they’ve had a chance to figure out what AI can actually do.

One global client we work with illustrates the shift I’m talking about. The company started its AI journey with a goal of saving roughly $1 billion. But as the work progressed, leaders realized that framing AI primarily around cost reduction was limiting both the ambition of the effort and the way employees engaged with it. The company realized it needed to change its approach.

Its revised goal? To create a culture of continuous AI innovation, a more inclusive and expansive approach that puts the focus on people and the future rather than on P&L statements. The saving will come as a trailing indicator of how well you have prepared your team. This shift has a lesson for all AI-integration projects: Don’t let chasing short-term profits distract you from finding out what your company could look like in the future.

And don’t make cost savings the only message employees hear about AI. Start instead by telling your employees that you’re embarking on this exploration together.

This changes the job for leaders. If we’re asking people to experiment and help us discover what AI can do, we can’t pretend we already know exactly where it will take us. Some experiments will work. Others won’t. Some roles will change in ways we didn’t anticipate.

Our employees can handle that uncertainty if we’re honest about it. What they need from us is not a promise that nothing will change, but a clear sense that we are investing in them and they will lead what comes next.

2. Trust the people closest to the work

The future version of your company probably isn’t going to come from the executive team. And that’s a good thing.

Big technology disruptions have historically been top-down. Leadership decided what the organization needed, rolled out the technology and trained employees to use it. The goal was adoption.

AI is different in that it gives us the chance to ask much bigger questions: If we were building this company today, knowing what AI can do, what would we build differently?

If we simply layer AI onto our existing processes, workflows and assumptions, we’re thinking too small. We may make yesterday’s business faster, but we won’t create tomorrow’s business.

So, who figures out what tomorrow’s business looks like? The people closest to the work. They know where customers get frustrated, where bureaucracy gets in the way and which parts of the job could be fundamentally different. They’re often in the best position to see what should be reinvented, not just what should be automated.

That means the message from leadership needs to change: We need your help imagining what’s possible. Give employees room to experiment, with IT providing the guardrails that keep the company safe without becoming a barrier.

And give them a voice in what shouldn’t change. Not everything needs to be automated.

3. Make the future a two-way investment

If we’re asking employees to help us imagine what’s possible, we have to invest in helping them get there.

The message from leaders should be simple:

• Get really good at the job you have.
• Discover what’s possible with AI across your work.
• Lead the change.  

If you accomplish these three objectives, demand for your talents will increase dramatically, and you will be part of creating the future of our company.

That’s the bargain. Companies need to provide the tools, time, coaching, workshops and transparency people need to adapt. Employees need to bring curiosity, experimentation and a thirst to keep learning.

One of my responsibilities as CEO is to make sure our people are ready to create what’s next. That’s good for the company, but it’s also one of the most valuable investments we can make in our people. Their jobs will change. Mine already has. The question is whether we see that change as a threat to manage or an opportunity to help our people develop new ways of working that are more valuable, meaningful and driven by uniquely human ingenuity.

The story you tell will shape what happens next

The way we approach AI will shape what our people do with it.

If people have outsiders impose AI solutions on them, they will see AI as a threat to protect themselves from, they’ll naturally look for ways to undermine AI and preserve the work they know. If they see it as an invitation to develop valuable skills and rethink their work, most will become the source of ideas to move the company forward.

At my own company, there are three messages I’ll keep repeating, as loudly and as often as I need to:

• Our AI journey will be shaped by our people—you are the experts in our business.
• We will invest to teach you the skills you need and will reward courageous learning.
• New innovations will lead to new roles.
• We will be transparent about what’s changing.

That’s what I believe leadership looks like in an AI-powered world.