AI has changed the context in which leaders operate, but it has not changed the basic work of leadership. People still need to understand where the organization is going, why it matters, and what’s expected of them. What AI adds is speed, uncertainty, and a technology that many employees are simultaneously curious about, afraid of, and already using.

As I’ve helped companies navigate their AI transformation journeys, I’ve begun to notice six things the strongest leaders do. The best part is that none of them require being the most technical person in the room.

1. Communicate ten times more than you think you should. In "Eight Reasons Why AI Transformation Efforts Fail," I revisited John Kotter’s warning that leaders tend to undercommunicate a transformation vision by a factor of ten. That feels especially true with AI.
   
   Leaders often assume their teams understand what they want because they said it once in an all-hands meeting or put it in their quarterly email newsletter. But it probably didn’t. You need to keep explaining why the work matters, what the organization is trying to achieve, what will change, and what will not. If you feel like you’re repeating yourself, you’re on the right track. Repetition is your friend.
   
   

2. Have a vision you can explain to anyone. Develop a thoughtful AI vision tied to the organization’s actual objectives. Then create the short version: one or two sentences you can remember and say to anyone at any time.
   
   A vision is only useful to your organization if people understand what it is and if they can remember it. So when you're communicating ten times more than you think you should, be sure that your vision is one of those things you're communicating.
   
   

3. Learn in public. No one has AI completely figured out. Even the consultants. And yes, even the forward-deployed engineers. Pretending otherwise creates alienation at exactly the moment when leaders need participation.
   
   Show people what you are trying, share what you are learning, and talk about the things that aren't working. Be willing to teach, but also be willing to be taught.
   
   

4. Make learning visible and social. You need to make this fun. Hold demo days, hackathons, or AI show-and-tells. Designate champions. Give out actual prizes. Create reasons for people to share useful experiments instead of keeping them hidden on their laptops.
   
   Gamification shouldn't be a substitute for strategy, but it can definitely lower the barrier to participation and make learning feel like something the organization is doing together. That's what you want. You also want to reward usefulness. When people see that, they get excited.
   
   

5. Go on a listening tour. This is one of those phrases you've probably heard before, but most leaders never really do it. I think it's actually one of the most useful things you can do as a leader. Ask your people how they’re using AI, how they wish they could use it, and what makes their work harder than it should be.
   
   Ask where they think AI could help, where it could fail, and what concerns them about the transformation. Ask them to give you specific examples. Figure out what they're excited about and what they're fearful of.
   
   The people closest to the work understand its friction better than anyone else. They can often see where AI could create real value and where an attractive idea will fall apart under the realities of the business. Listen carefully, and then act on some of what you hear.
   
   

6. Give people time and permission to experiment. Leaders sometimes announce that AI is a priority and then expect people to learn it in whatever time they have left over. That’s not a transformation plan. You've got to support people in the effort, and that's going to mean giving them time to do the things you want them to do.
   
   Give people approved tools, clear guardrails, protected time, and a way to bring what they learn back to the organization. Make it clear that a well-run experiment can be valuable even when it disproves an idea. If you want people to learn, you have to make learning a legitimate part of the work.
   
   Leading in the AI era doesn’t require having every answer or knowing every tool. It requires creating clarity, making learning safe, and giving people a practical way to turn curiosity into useful change.
   
   The technology will keep moving. Your responsibility as a leader will not.