AI: The Amazing Intern
AI can create a lot of opportunity. But it can also create a lot of anxiety.
That anxiety often shows up as a quiet worry that everyone else has figured something out you haven’t.
If you can relate to that, this is for you.
Having taught, coached, and worked with AI-curious folks from around the world, I see one mindset that helps reshape that anxiety into confidence.
It’s not technical. It’s not a prompt formula.
It’s a way of thinking about what AI actually is.
Most of us were quietly handed the idea that AI is software we’re supposed to master.
That framing alone is enough to make anyone freeze.
What if it isn’t software you have to master? What if it’s more like an amazing intern?
Capable. Eager. Available.
Still needing your direction, your context, and your feedback.
That one shift in how you see it can change what you expect, how you approach it, and what you get back.
The Amazing Intern
Think about what you’d actually do if you brought a real intern into your business.
You wouldn’t sit them down on day one, say “create something useful for me,” and walk away. You’d give them background. You’d explain what you’re working on and why it matters. You’d show them what a good result looks like. You’d expect questions. When they brought back something that was close but not quite right, you’d give feedback. “This part is good. Make this section more specific.” “Try again, but shorter.”
That same posture is what works with AI.
Think about what happens when you type “write me an article.” AI has no choice but to guess. An article about what? Who’s reading it? What do you want them to feel or do by the end? What’s your point of view? Should it sound like you, or like someone else? It doesn’t know unless you tell it. So it gives you the average of everything it has ever seen on the topic, which is exactly the generic answer you walked away from.
When you treat it like an intern, you naturally give it more of what it needs. You explain the situation. You give context. You describe what good looks like. You guide the work.
AI is not magic. It’s not a mind reader. It’s not the boss.
It’s a capable helper that does surprisingly good work when you lead it well.
And by following a few simple frameworks, you can do exactly that.
Start With One Thing
One of the quickest ways to feel stuck with AI is to try to use it for everything at once.
You read an article that says it can write your emails, build your strategy, summarize your meetings, brainstorm your campaigns, clean up your data, and replace half your team. So you sit down, open the tool, and freeze. Where do you even begin?
Start with one thing.
Not your whole business. Not your whole job. Not some giant process that touches every part of your day.
One thing.
Something you already do that takes time, feels repetitive, or makes you think, “there has to be an easier way.” A first draft of an email. Rough notes you need to clean up. A document you need to pull main points from. A messy idea you need to think through before a meeting.
This is where I see AI earn its place fastest. So much of your day gets eaten by what I call the busy middle. The mechanical work that crowds out your actual thinking. AI is good at contracting that middle. Done well, it gives you back room to do what only you can do.
The task doesn’t have to be impressive. It just has to be real.
You can even say to AI:
“I do this task a few times a week, and it takes more time than I’d like. I’m not sure how you can help yet, but I’d like to figure that out. Can you give me a few ways you might be useful?”
That’s a low-pressure way in. You don’t have to know how AI can help before you ask. You’re turning it into a problem-solving partner instead of an order-taker.
Once you see it help with one real thing, something shifts. You start understanding the tool through your own experience instead of through someone else’s headline.
That’s where confidence starts.
Talk To It Like A Person, Not A Computer
If there’s one shift that changes everything about how AI feels to use, this is it.
Most of us have spent twenty years using Google. You type a few words and hope it points you toward the right place. You search. You scan. You click. You compare. We’ve all been trained to be efficient with our words because the tool didn’t reward anything else.
AI works differently. AI gives you the chance to have a conversation.
That first time you tried it, you probably brought your Google habits with you. A few words. A quick question. Hit enter. That’s not a mistake. It’s just years of muscle memory. It’s also why the answer came back flat.
The good news is you already know how to do this. You explain things to people every day. You tell your team what you need. You walk a customer through a question. You think out loud with a colleague. That’s all this is.
You can ramble a little. You can say what you’re trying to do, why it matters, what feels hard, where you’re stuck. You can say, “I don’t know exactly what I need yet, but here’s what I’m trying to figure out.” That kind of honesty is exactly what helps.
Here’s what that might sound like. Imagine you’re a marketing director who spends a lot of time writing customer emails. You could open AI and say:
“I’m a marketing director, and I spend a lot of time writing emails to our customers. It matters because those emails are often the only direct touch we have with them between purchases. I’d like to make this easier and more consistent without losing the warmth. How could you help?”
That’s not a fancy prompt. That’s a conversation.
Once AI gives you ideas, the first answer isn’t the finish line. It’s an opening to keep going.
“That second idea sounds interesting. How would that work in my situation?”
“I like the direction. Make it less formal.”
“Can you give me an example?”
“Ask me a few questions before you suggest anything else.”
The first answer isn’t the whole experience. It’s the start of the conversation.
The Four Things To Give It Before You Ask
Once you start having real conversations with AI, you’ll notice the answers get better the more you give it up front.
I think about it as four things. The same four things you’d give a person if you wanted them to help you well.
1. What we’re doing
The actual task. Not the topic, but the specific thing you’re trying to get done.
There’s a real difference between “help me with marketing” and “I’m trying to write a short email to past customers inviting them to schedule a follow-up call.”
The second one gives AI something clear to work on.
2. Why we’re doing it
The reason behind the task. The thing that makes this matter to you.
“I’m sending this because we haven’t talked to these customers in a while, and I want the email to feel helpful instead of salesy.”
Now AI understands more than the assignment. It understands the purpose. That helps it make better choices.
3. What success looks like
A picture of what a good answer should look like when it’s done.
“Success looks like a short, friendly email that sounds like it came from a real person. Warm. Simple. Easy to reply to.”
Now there’s a target. Not just “an email.” The kind of email you actually want.
The same idea works almost anywhere:
“Success looks like a bulleted list I can bring into a meeting.”
“Success looks like a plain-English summary I can send to someone who doesn’t know the background.”
“Success looks like three practical ideas I could try this week.”
The clearer the finish line, the better chance AI has of getting you there.
4. Any questions for me?
This is the one most people skip. It might be the most useful.
After you’ve shared what you’re doing, why, and what success looks like, you simply ask:
“Do you have any questions for me before you start?”
That one sentence changes the dynamic. Instead of AI guessing what you mean, it has permission to slow down and ask for what it needs.
It might ask about tone. About audience. About length. About what you’ve already tried. Those questions often help you realize what you forgot to include.
Here’s what the whole thing sounds like put together:
“I’m trying to write a short email to past customers inviting them to schedule a follow-up call. I’m doing this because we haven’t talked to them in a while, and I want it to feel helpful instead of salesy. Success looks like a short, friendly email that sounds like it came from a real person and is easy to reply to. Do you have any questions for me before you start?”
Simple. Clear. A much better chance of useful work.
Stay In Charge
Here’s how I think about your role in all of this.
Your value isn’t writing the first 800 words. Your value is making those 800 words as good as they can be. AI can help you get to a first version faster, which gives you back the time and energy for what actually matters: your judgment, your taste, your call.
That’s what staying in charge looks like. It’s not about distrust. It’s about ownership. The work still has to pass through you before it goes out into the world.
The reason this matters is that AI sounds confident even when it shouldn’t. It writes quickly. It organizes ideas neatly. It gives you something polished enough to trust.
Polished doesn’t always mean right.
You’re the one who knows your customer. You know your team. You know the small details that make something feel right or wrong. AI doesn’t have any of that unless you tell it, and even then, the work still has to come back through you.
Sometimes that means correcting it. The same way you’d give an intern feedback when they brought you something close but not quite there:
“This is close, but it sounds too polished. Make it sound more like something I would actually say.”
“The structure is good. The examples are too generic. Ask me questions so you can make this more specific.”
“I like the idea, but I don’t fully trust the details. What should I double-check before using this?”
That last one matters most when accuracy is on the line. Facts, numbers, names, quotes, policies, legal details, medical advice, financial advice, anything where being wrong has consequences. AI can help you with all of those. It just shouldn’t be the final word on any of them.
That’s not a reason to avoid AI. It’s a reason to use it wisely. The judgment, the taste, the relationships, the goals, the real-world understanding of what matters: those are still yours. AI just helps you get there faster.
Confidence, Not Mastery
The goal of all this isn’t mastery.
Not understanding every feature. Not keeping up with every new tool, every headline, every opinion flying past you. That race is exhausting and most of it doesn’t matter to your day anyway.
The goal is to get confident enough to stay curious.
Here’s what I see with clients. When you feel intimidated, you freeze. When you feel behind, you avoid it. When you think you have to know everything before you start, you never start.
A little confidence changes that. You ask a question. You try one task. You give more context. You ask AI to explain something another way. You watch it actually help with something real, and the whole thing gets less mysterious.
Little by little, AI stops feeling like something happening somewhere else, to someone else, and starts feeling like a tool you can actually use.
Once you find one useful thing, the next one shows up. AI helps you write an email faster. Then it helps you clean up meeting notes. Then it helps you prepare for a customer conversation. Then it helps you turn a messy idea into a clear outline.
Each small win gives you a little more confidence. Each bit of confidence makes it easier to stay curious.
Remember the worry we started with. The quiet feeling that everyone else has figured something out you haven’t. That worry doesn’t disappear because you’ve caught up. It disappears because you’ve stopped trying to.
You don’t have to catch up. You don’t have to know everything. You just have to be willing to try again, with a better way to think about it.
One conversation at a time.