AI Isn’t Going to Replace You. But Letting It Think for You Might.

Let’s face it. The world has changed.

Three years ago, artificial intelligence was a concept most people associated with science fiction, Silicon Valley, or some distant version of the future. Today? It’s practically a member of the household. We’re asking ChatGPT to help us navigate everything from complex business decisions and workday responsibilities to home renovations, parenting challenges, and dating advice. You name it, there’s probably an AI prompt for it. And I get it. The technology is remarkable, dare I say addicting.

We’ve gone from typing a few keywords into Google and sifting through pages of search results to having what feels like an intelligent conversation with a tool that can synthesize information, contextualize questions, and deliver beautifully articulated responses in seconds. It’s an extraordinary leap forward. But somewhere along the way, I think we’ve started blurring an incredibly important line.

There’s a difference between using AI to help you think and allowing AI to think for you.

And in the workplace, that distinction might be more important than ever.

The Question Isn’t Whether You Use AI. It’s Who’s Leading.

There’s no shortage of conversations about AI replacing jobs. And while I believe some of those concerns are entirely warranted, I think we’re overlooking a more immediate risk. What happens when we voluntarily hand over the very things that make us valuable as professionals? Our critical thinking. Our experience. Our judgment. Our ability to connect dots, challenge assumptions, and recognize when something simply doesn’t make sense. Think about it.

AI can analyze a hundred-page document in seconds. It can summarize research, write code, organize data, develop presentations, and turn a rough collection of notes into something that looks remarkably polished. Tasks that once consumed hours, sometimes days, can now be accomplished in a fraction of the time. That’s an incredible opportunity. But efficiency and expertise are not the same thing.

Just because AI can produce something doesn’t mean it understands the problem it’s supposed to solve. And just because an output looks impressive doesn’t mean the thinking behind it is sound. Which brings me to a distinction I believe every professional should understand…

 

AI-Led vs. Human-Led: There’s a Difference

I like to think about AI in the workplace through two approaches: AI-Led and Human-Led. 

1. AI-Led thinking happens when we start with the tool and allow it to determine the direction. We ask AI what we should do, what we should recommend, how we should approach a problem, or what our opinion should be. We accept its initial framing, build upon its recommendations, and gradually allow its output to become the foundation of our own perspective. Sure, we might make some edits along the way. But the thinking itself originated somewhere else.
2. Human-Led thinking, on the other hand, starts with us. We bring the experience, the context, the hypothesis, the curiosity, and the point of view. We use AI to explore possibilities, pressure-test assumptions, accelerate execution, and bring our ideas to life. The difference isn’t how much AI you use. It’s how much authority you give it. Let’s use a simple workplace example. Imagine you’re a consultant tasked with assessing a client’s marketing operations.

Both approaches use AI. Both might produce a beautiful presentation. But only one guarantees that a human has intentionally shaped the thinking behind it. And when the client asks, “Why are you recommending this?” that difference becomes painfully obvious.

Your Expertise Is the Asset. AI Is the Accelerator.

Here’s where I think we need to fundamentally rethink our relationship with this technology. AI should be a tool in your toolkit, not the person holding the tools. Think about a carpenter, an artist, a doctor, or a consultant. A carpenter doesn’t become valuable because they own a hammer. Their value comes from understanding what they’re building, knowing which tools to use, anticipating problems, and making decisions that ultimately determine the quality of the finished product.

The same principle applies to AI. You are the subject matter expert. You are the one who understands the business, the client, the audience, the history, and the nuances that don’t always make it into a prompt. AI can help you execute your ideas faster, but it shouldn’t be responsible for deciding which ideas are worth pursuing in the first place. And here’s something else worth acknowledging: you don’t always have to know the answer.

Being Human-Led doesn’t mean being the smartest person in the room or having an immediate solution to every problem. It means taking responsibility for finding the right answer. Sometimes that means asking better questions. It might mean challenging your own assumptions. Sometimes it means admitting you don’t know enough yet and doing the work to understand. AI can absolutely support that process. In fact, it can make us better at it. But curiosity, accountability, and the willingness to exercise judgment? Those responsibilities still belong to us.

So, What Does Human-Led Actually Look Like?

I don’t believe the answer is to use AI less. Quite the opposite. I believe we should be using it more intentionally. A few principles I try to keep in mind:

1. Think before you prompt: Before asking AI to solve something, ask yourself what you already know. What’s your initial perspective? What problem are you actually trying to solve? What would a successful outcome look like? Even a rough hypothesis gives you something to work from, rather than something to blindly accept.
2. Feed AI your thinking, not just your assignment: There’s a significant difference between asking, “Create a strategy for this client,” and saying, “Here’s what I’ve observed, here’s what I believe the underlying problem is, and here’s the direction I’m considering. Help me pressure-test it.” The second approach positions AI as a collaborator rather than the decision-maker.
3. Use AI to actualize your ideas: Have a concept but aren’t sure how to bring it to life? Great. That’s where AI shines. Use it to build, structure, visualize, analyze, and execute. Let it take on the heavy lifting so you can spend more time on the work that requires your expertise.
4. Challenge the output, even when it looks incredible: This might be the most important principle of all. AI is exceptionally good at making things look and sound convincing. A beautifully formatted presentation. A compelling executive summary. A confident recommendation. But we’ve all heard the saying: Don’t judge a book by its cover.

The same applies here. A polished output can contain flawed assumptions, missing context, misinterpreted information, or recommendations that sound reasonable but simply aren’t right for the situation. Ask yourself: Do I agree with this? Can I defend it? Does it reflect what I actually believe? And if the answer is no, don’t let the polish convince you otherwise.

5. Remember who’s accountable: At the end of the day, your name is on the deliverable. Your reputation is attached to the recommendation. Your client, colleague, or leader is trusting your expertise. Not your AI subscription.

Don’t Confuse a Professional-Looking Output With Professional Judgment

I think this is where the greatest danger lies. We’ve become so accustomed to AI producing something impressive in seconds that it’s easy to assume the hard work is finished. But what if the hard work was never the writing, formatting, or production in the first place? What if the real value has always been in knowing what to say, why it matters, and what should happen next? That’s the work we cannot afford to abandon. In fact, I would argue that as AI makes execution increasingly accessible, the ability to think critically, exercise sound judgment, and bring a differentiated perspective becomes even more valuable.

Anyone can ask AI to generate a recommendation. Not everyone can recognize whether it’s the right recommendation. Anyone can produce a polished presentation. Not everyone can walk into a room, defend the thinking behind it, navigate difficult questions, and adjust their perspective when new information emerges. That’s where human expertise earns its place. And it’s precisely why the next step after Human-Led thinking is learning how to critically evaluate what AI produces.

In the second part of this series, my colleague Ali Barnard explores that challenge in Editing AI: What Accurate Output Still Gets Wrong, including why factual accuracy isn’t enough, how polished AI outputs can disguise poor judgment, and what it really takes to make AI-generated work client-ready.

The Future Belongs to People Who Keep Thinking

I want to be clear about something. AI will change jobs. Some roles will be eliminated, others will evolve, and entirely new ones will emerge. Pretending otherwise would be naive. But I don’t believe the answer is to compete with AI on speed, volume, or production.

That’s a race we’re unlikely to win. Instead, I believe our greatest opportunity is to double down on the qualities that make us valuable in the first place. Our experience. Our creativity. Our intuition. Our empathy. Our ability to navigate ambiguity, build relationships, and make decisions when the answer isn’t obvious. And perhaps most importantly, our willingness to think for ourselves. So the next time you open ChatGPT, Claude, or whichever AI tool has become your personal favorite, consider asking yourself one question before you type:

Am I using this tool to make my thinking better, or am I asking it to do my thinking for me?

Because AI is an extraordinary accelerator. But you should still be the one behind the wheel.