AI: like it, hate it, or still don’t quite get it, you can’t avoid it.

I’ve used AI since ChatGPT came out. I try to pay attention to how it’s affecting business and society. I read the projections. Some utopian, some apocalyptic.

Recently, I attended an AI training on how to build a team of AI agents, managed by one AI agent who reports directly to me. Throughout the training, I kept thinking about how using AI well is a lot like leading well.

You might call the overarching theme: ‘Garbage in, garbage out.’

Within the first few minutes, my mind was blown by the possibilities. About ten minutes later, I had flashbacks to a college lab where everyone else was furiously typing and I still couldn’t get past the blinking cursor.

But besides my own learning curve, one thing stood out: you have to know what you want, and you have to be able to explain it.

AI can do amazing things. It can also go in crazy directions. The more vague you are about your desired outcome, the more likely it is to produce slop. If you want an AI agent to work autonomously, you have to define the result you want it to achieve: what it’s supposed to accomplish, what success looks like, how it will know it’s done.

You also need guardrails: what it should and shouldn’t do while pursuing the objective. Otherwise, it creates new problems along the way.

Exactly like great leadership.

Leaders often feel frustrated by their teams. They feel like people constantly need approval, input, or direction, or can’t meet expectations or work within standards. But the larger your team gets, the more you have to delegate, and the more your leadership quality gets amplified through the system.

If you know and communicate what you want, how you’ll measure results, and the guardrails around getting there, you can build a powerful, aligned organization. Anywhere you haven’t defined those things, problems will almost certainly emerge.

Building a team that can work without you in the weeds means doing your own work first:

• Being clear about your expectations, starting with your vision and priorities
• Actually using your values, policies, and SOPs as guides and guardrails.

The more of that leadership work you do upfront, the more independently and successfully your team can work toward the organization’s goals.

The discipline required to be effective with AI translates directly into being more effective leading humans. Even if you have zero interest in AI, it’s a useful leadership sparring partner. It tests the same fundamental skills, just with faster, lower-stakes feedback loops.

Simple. Not easy.

Leadership comes down to three things:

• Defining an ideal future or end state
• Building and aligning the people who will create it
• Clarifying and protecting the standards that guide them along the way.

Some readers will look at those three and say, ‘Leadership is more difficult than that.’ Of course it is. But difficult and complicated aren’t the same thing.

Most leadership challenges aren’t complicated to answer. They’re just difficult to do. Mostly because they feel difficult, not because they actually are.

If you’re a leader, your real job may be far less complicated than you’ve made it. Develop, communicate, and reinforce clarity, and the job simplifies radically.

I’ve coached executives and owners for about twenty years. Most of them can go into great detail, off the cuff, about their business performance and the internal dramas on their teams. They know their customers, their people, their problems, their history.

But ask them these three questions, and most struggle:

1. What exactly are you trying to build, and how will you know you’ve built it?
2. Do you have the right people in the right positions, and how are you training and supporting them to succeed?
3. Does everyone understand the standards for how you work and make decisions, and are those standards practiced consistently?

None of these questions are technical or complicated. But few leaders can answer them clearly. That kind of clarity takes work and courage. And those are the two places this tends to fall apart.

• Leaders don’t put in the upfront effort to define or communicate what they want.
• Fear of accountability or pushback holds leaders back.
• (I know I said two, this is a bonus) Many leaders also aren’t sure what they really want or expect for themselves.

Try this for the final quarter of the year. Develop clear answers to these questions, scoped to your goals for Q4:

• What is the one outcome that matters most?
• Who needs to be aligned around achieving it, and what support do they need?
• What standards or guardrails do you need to define or protect as you pursue it?

Then spend the rest of the quarter communicating, reminding, checking progress, adjusting, and supporting your team’s success.

Test whether it doesn’t change your outcomes.

AI is changing how we work. But it is magnifying the need for leadership. Before you can effectively lead people toward an outcome, or build an AI agent, you need to know exactly where you’re asking them to go.

That clarity also determines whether a business can eventually run, scale, or sell without you trapped in the day-to-day. If you’re working through what that looks like for yours, reach out.

Take good care,
Christian