How do you actually get value from AI as a leader?

This question comes up quite a lot in my AI leadership advisory sessions. Most of the leaders I work with are already using AI. They use it to write, summarise, search for information or get quick input on a task. And it can certainly save time.

But I think it becomes more interesting when we start talking about what the value is actually supposed to be. Because if the goal is simply to produce the same email a little faster, we are only tapping into a small part of the potential.

For me, it makes sense to look at the value on three levels: What does AI do for you as a leader? What does it mean for the people you lead? And when does it begin to create value for the organisation?

AI should do more than save you time

Time is, of course, valuable. If AI can help you create an overview, follow up on things or complete a task in half the time, there is real value to be gained. I also find that it can create a sense of relief when you do not have to keep every task and follow-up in your head yourself.

But I think the more interesting value emerges when AI helps you improve the work itself. That might be when you use AI to challenge your thinking on a strategic issue, prepare for a difficult conversation or gain different perspectives on a decision.

One of the things I work on in my advisory sessions is therefore bringing AI into the process earlier. Many people start by writing a text, creating a presentation or thinking through most of a strategy, and then ask AI whether it can improve the result. But why wait until the end? If AI is going to improve the quality of your work, it is often far more interesting to use it along the way to challenge assumptions, identify blind spots or explore other possibilities.

At some point, the question therefore shifts from “What can AI produce for me?” to “Where in my work can AI help me do a better job?”

When should we not use AI?

That may be just as important. Because in my view, using AI well is not about using it as much as possible. Quite the opposite.

I think AI only really becomes part of your leadership practice when you approach a task and naturally consider whether AI should be involved – and how. But also when you are able to decide that, in this particular case, it should not be involved at all.

That requires judgement. AI can be very useful when preparing for a difficult conversation, but you are the one who has to have the conversation. It can challenge a strategy, but you need to understand and own the choices that end up in it. And it can help with a first draft, but if your name is on it, you also need to be able to stand behind the content.

This is one of the boundaries I find myself thinking more about: You can give AI work to do. But as a leader, you cannot hand over your responsibility.

The value does not stop with you

As a leader, your own use of AI is only one part of the task. You also lead employees who are already using AI in very different ways – and some may not be using it at all.

Here, I think the leadership task is different. It is not your job to identify every good use case or tell employees exactly how they should use AI. But you do have a responsibility to create the conditions in which they can figure it out.

Do employees know what they are allowed to use AI for? Is it clear what you expect? Is there room to experiment and talk about what did not work? And do you share the useful experiences, or do they remain with the individual employee?

Of course, this also requires knowing what is already going on. There are probably employees who have found genuinely useful ways of working with AI, others who are still uncertain, and perhaps also uses of AI that you, as a leader, need to address. That is why I think we need to start talking more concretely about AI, both within teams and leadership groups – not just about the tools, but about what we are actually using them for and what we are learning from it.

When does AI become organisational value?

And then there is the organisational level. Because even if 100 employees are each saving time by using AI, that does not necessarily mean we have identified its greatest potential.

Perhaps the more interesting value does not lie in making all our existing tasks a little faster. Perhaps it lies in looking at the work itself and asking whether some of our tasks and workflows should look different now that AI is available to us.

I think that is an important distinction.

If AI only comes in at the end and helps us write a document faster, we have made one part of an existing process more efficient. But what if AI could be used earlier? What if it could change the information we collect, who needs to be involved, or how the task is approached in the first place?

At that point, AI becomes something more than an individual productivity tool. And that also requires more than one employee with a good idea. We need to be able to identify the uses that genuinely have potential, involve the relevant expertise and determine whether what we develop actually creates value in practice.

This is closely connected to what I have previously written about organisational AI maturity. Because if the experience stays with the individual, the organisation does not necessarily become much wiser – no matter how skilled individual employees become at using AI.

So for me, getting value from AI as a leader is increasingly about more than becoming good at using the technology yourself.

It is, of course, about whether AI can help me do my own work better. But it is also about whether I create the conditions for employees to use it well. And whether, as an organisation, we become good enough at identifying the places where AI can do more than make existing work faster – where it might actually lead us to work differently.

That, at least, is where I would start looking for the value.

/Julie