You don’t need to understand artificial intelligence to start using it.

You don’t need to learn “prompt engineering.”

You don’t need to become a technology expert. You don’t need lessons.

And you certainly don’t need to reorganize your life around another piece of software.

If you’ve been curious about AI but haven’t quite understood what all the excitement is about, I suggest something much simpler:

Have five conversations with it.

Not five computer exercises.

Five conversations.

For this experiment, I’m going to use ChatGPT because its voice feature allows you to simply talk—and because it can remember useful preferences so you don’t necessarily have to introduce yourself over and over again.

First, get your voice ready

Open ChatGPT on your phone.

Look for the Voice icon in the message area and tap it. The first time you use Voice, ChatGPT may ask permission to use your microphone and allow you to select a voice.

Don’t agonize over the choice.

Listen to a few and pick one that you find pleasant to talk with. You can always change it later under Settings → Voice.

Then forget about the controls for a moment.

Just talk.

You can speak naturally. You can interrupt. You can ask a follow-up question. You can tell it to slow down, shorten an answer, explain something differently, or go into more detail.

That’s the first thing to understand:

You aren’t operating it. You’re conversing with it.

Conversation #1: Start with something you already know

This may seem backward.

Why ask AI about something you already understand?

Because right now you’re not testing your knowledge.

You’re testing the AI.

Try:

“I’m new to AI. Let’s talk about something I know fairly well. Explain why strength training becomes more important as we age. Give me the three main reasons.”

Listen to the answer.

Then challenge it:

“What evidence supports that?”

Or:

“I’m not convinced about your second point. Explain it.”

Or simply:

“That’s too complicated. Say it more simply.”

Why start here?

Because when you already understand the subject, you can judge whether the AI makes sense.

That’s a much better first experience than asking it something important and simply hoping the answer is correct.

Conversation #2: Ask it to explain something you’ve wondered about

Now give it a subject that interests you but that you’ve never completely understood.

For example:

“I keep hearing the term VO2 max. Explain what it means to an intelligent person who isn’t a physiologist.”

Then keep going.

“Why should someone my age care about it?”

“What actually improves it?”

“Which claims about it are well established and which are still uncertain?”

Why do this?

Because this is where AI begins to feel different from searching the internet.

Instead of reading ten articles and trying to piece them together, you can keep asking questions until the explanation makes sense to you.

Conversation #3: Don’t ask for an answer. Ask it to interview you.

This is one of my favorite ways to introduce someone to AI.

Say:

“I’m considering walking, cycling and swimming as ways of improving my cardiovascular fitness. Don’t give me an answer yet. Ask me one question at a time and help me think through which combination might suit me best.”

Then answer its questions.

Let the conversation develop.

Why is this important?

Because AI isn’t limited to answering questions.

It can also help you think through a decision.

That is a very different relationship.

You’re no longer asking:

“What does the computer think I should do?”

You’re saying:

“Help me think.”

Conversation #4: Use it to prepare for a real conversation

Suppose you have an annual physical coming up.

Try:

“I have an annual physical coming up. Help me prepare questions I might want to discuss with my doctor about strength, balance, cardiovascular fitness, sleep and preventive health. Don’t diagnose me or tell me what treatment I need. Help me prepare for a better conversation with my physician.”

You could do the same thing before meeting a financial adviser, taking a trip, buying a computer, talking with a contractor or starting a new hobby.

Why?

Because AI can be very useful before an important decision without being the one making the decision.

Think of it as preparation.

Not authority.

Conversation #5: Challenge the AI

This may be the most important exercise of all.

Ask something controversial or confusing.

For example:

“Give me the case for and against taking a daily multivitamin after age 65. Separate well-established evidence from uncertain claims. Tell me where you’re uncertain and give me reputable sources I can check myself.”

Then push back.

“Are you sure?”

“What is the strongest evidence against that position?”

“What would someone who disagrees with you say?”

“Which part of your answer are you least confident about?”

Why?

Because you should never learn to trust AI blindly.

You should learn to question it intelligently.

A good AI user isn’t someone who believes everything the machine says.

A good AI user knows how to make the machine explain itself.

Now do one more thing: introduce yourself

Here’s where we can remove one of the biggest annoyances of using AI.

You shouldn’t have to explain yourself from scratch every time you open ChatGPT.

So don’t think of this as “configuring your system.”

Think of it as:

Introducing yourself to your AI.

You might simply say:

“Before we continue, I’d like you to understand how I prefer to work with you.

I’m retired and interested in staying healthy, active and independent. I’m curious and I like understanding the reasoning behind things, but I don’t want unnecessary jargon.

When we’re discussing important subjects, especially health, tell me when something is well established and when the evidence is uncertain.

Don’t simply agree with me. If you think I’m overlooking something, tell me.

Keep your first answer reasonably concise. I can always ask you to go deeper.

And when appropriate, give me reliable sources so I can check important information myself.”

Then add:

“Please remember these preferences for future conversations.”

That’s it.

You have just done something that sounds terribly technical when computer people describe it.

You’ve personalized your AI.

ChatGPT’s Memory feature can retain useful information from conversations when memory is enabled, and you can review, change or delete remembered information under its Personalization settings.

You can also ask:

“What do you remember about me?”

And if something isn’t right:

“Forget that.”

Or:

“Remember this instead.”

You remain in control.

One final rule

Don’t begin your AI journey with something terribly important.

Don’t start by asking it to diagnose an illness.

Don’t make a major financial decision because it told you to.

Don’t assume that because an answer sounds confident, it’s correct.

Instead, spend a little time discovering what it’s good at.

Talk to it.

Question it.

Interrupt it.

Correct it.

Ask it to explain itself.

Ask it to show you its sources.

And perhaps most importantly:

Disagree with it occasionally.

You may discover that learning to use AI doesn’t require learning much about computers at all.

It requires something you’ve been doing your entire life:

Having a good conversation.

Start with five.

Then decide for yourself whether the sixth one is worth having.

/