My Teammate Told Me He Was Becoming a Vampire

Recently I got on a video call with one of my most senior teammates. He had a serious look on his face. The kind of look that makes you think someone is about to tell you something important.

Then he said:

Dead serious.

I just stared at him. He stared at me. I remember thinking, Am I in a dream? Is this actually happening? What am I supposed to say to that?

Finally I managed: "Okay. What do you mean?"

Thankfully, he was not announcing a sudden interest in coffins, capes, or drinking blood. He was talking about AI. And that conversation taught me something I think a lot of leaders are still missing about what happens when talented people really start working with AI.

The AI Vampire

My teammate was using AI heavily in his work. Not occasionally. Not asking it a question here and there.

He was creating with it, directing it, checking its work, giving it more instructions, reviewing the output, changing the prompt, starting another process, then another. And he was loving it. The possibilities were exciting. The productivity was exciting. The amount he could create was exciting.

There was always one more thing he wanted to try. One more prompt. One more change. One more idea. One more output to check.

He even told me about bringing his computer with him to his daughter's hockey practice and having her watch it. "Tell me when the prompt is done. Tell me what it says."

I laughed. Then I realized how serious the conversation actually was.

Because he really was becoming a vampire.

He was staying up later. He was constantly checking what the AI was doing. His brain was continuously engaged with what could be built next. And even though he was enjoying it, it was also burning him out.

Both things were true at the same time.

Excitement Does Not Protect People From Burnout

I think this is going to become an important distinction as AI becomes embedded in more jobs.

We traditionally think about burnout as something caused by work people dislike. Too many meetings. Too much administrative work. Unrealistic expectations. Not enough resources. Constant pressure. Those things contribute to burnout, and they are most of what we look for when we go looking for sustained overload on a cybersecurity team.

AI introduces another possibility.

That is what was happening. My teammate was not disengaged. He was not unmotivated. He was not complaining that his job was boring. He was deeply engaged.

He could suddenly create things faster than before. He could explore ideas immediately. He could accomplish work that previously would have taken more time or more people. And that created its own kind of pressure. Not pressure coming from me. Pressure coming from possibility.

• If I can do this, what else can I do?
• If the AI is still working, why would I stop now?
• If I can have this finished tonight, why wait until tomorrow?
• If one more prompt could make this better, why wouldn't I send it?

The AI does not get tired. The human does.

This is also why the signals leaders rely on are so easy to misread. We are trained to notice the person who has gone quiet. We are not trained to notice the person who cannot stop.

We Had to Talk About It

Once I understood what was happening, the conversation shifted.

This was not about asking him to be less creative. It was not about telling him to stop experimenting with AI. I did not want to extinguish the excitement. This was about figuring out how to harness that excitement without allowing it to consume him.

We had to talk about boundaries. About sleep. About shutting the computer. About allowing an AI process to wait until morning. About the fact that being capable of producing more does not mean a human being should be producing every possible minute. About self-care. About sustainability.

And perhaps most importantly, about recognizing that maximum productivity is not the same thing as healthy productivity.

That distinction matters.

AI Is Changing Work in Ways Leaders Cannot See

There is a lot of conversation about what AI will automate. Which jobs will change. Which jobs will disappear. How many hours AI will save. How much productivity companies will gain.

Those are important questions. I think there is another one leaders need to start asking.

What is happening to the people who are becoming very good at working with AI?

Because their work is changing too. Someone has to decide what the AI should do. Someone has to provide context. Someone has to evaluate what comes back. Someone has to redirect it. Someone has to decide what is right, wrong, useful, risky, complete, or ready. Someone has to connect all of that output to the business objective.

When someone becomes very good at doing this, the velocity of their work climbs fast. That can be exhilarating. It can also be exhausting. And almost none of it is visible from the outside.

This Is Bigger Than Engineering

A lot of this behavior is showing up first among technical people, because developers and product teams already use powerful AI tools throughout their workflows. It will not stay there.

• Marketing professionals
• Researchers
• Operations teams
• Salespeople
• Security and IT teams
• Executives
• Entrepreneurs

Anyone who moves from occasionally using AI to actively working alongside AI can experience this shift. I think organizations are underestimating what that transition asks of their people.

AI proficiency is not simply knowing how to write a prompt. It is learning how to manage a new flow of work. It is knowing when to keep going. It is knowing when the AI needs intervention. It is knowing when something is good enough.

And apparently, it is also knowing when to close the laptop and go to bed.

Leaders Need to Understand the Work

This experience reinforced something I have believed for years. You cannot manage a workforce by knowing people's titles. You have to understand the work.

What are people actually doing? What tasks are they responsible for? What projects are consuming their time? What capabilities do those projects require? What work is AI now doing? What work is the human still doing? What new work has been created because AI entered the workflow? And what is happening to the workload of the person managing all of it?

AI may reduce effort in one area and expand possibility in another. That is the part that never shows up on a headcount report, and it is why seeing work at the task and project level matters more now than it did before AI entered the workflow.

If leaders only measure output, they may look at someone like my teammate and think: This is great. Look how productive he is.

And it was great. He was productive. He was also exhausted. Both were true.

The Future of Work Still Has Humans in It

I do not want people burning themselves out in pursuit of AI productivity.

I want AI to give people greater professional efficacy. I want it to eliminate unnecessary work. I want it to expand what people are capable of creating. I want people to feel energized by what is possible.

And I also want leaders to remember something simple.

That means workforce management in the AI era cannot only be about getting more output. It has to include understanding workload, tasks, projects, capabilities, and the capacity of the people holding all of it together.

The workforce has already changed. AI is now part of the ecosystem doing the work, alongside employees, contractors, consultants, and MSPs. What has not changed is that a person still sits at the center of it, deciding what the work should be and whether it came back right.

My teammate is still using AI. He is still excited about it. He is just not doing it at two in the morning anymore.

That conversation only happened because he said something out loud, in a strange enough way that I had to ask what he meant. Most leaders will not get that. Most will get a person who looks like the best quarter they have ever had, right up until the moment they do not.

---

About CyberSN Workforce Intelligence

CyberSN's Workforce Intelligence Engagement gives cybersecurity and IT leaders a clear view of the tasks and projects their workforce actually performs, who and what is performing them, and where capacity, capability, and workforce risk sit today.

It measures capability coverage, capacity, maturity, redundancy, and dependencies rather than headcount. So when AI changes what the work is, leadership can see where that work actually landed, where sustained overload is forming, and which person looks highly productive while running past what is sustainable.