🌱 Still Nervous About AI? Ask It These Five Tiny Things

If you’ve been circling AI for months - reading about it, hearing about it at work, feeling that low hum of “I should probably get on with this” - I want to let you in on something I’ve learned from teaching a lot of nervous beginners: confidence doesn’t come from understanding AI. It comes from five tiny asks where nothing was at stake.

That’s it. Not a course. Not “getting your head around it” first. The people who went from anxious to comfortable in my sessions all did it the same way: they asked something small, nothing bad happened, something mildly delightful happened instead, and a week later they’d stopped thinking of it as A Big Thing.

So this is the gentlest guide on this site. Five asks, spread over a week or done in one sitting, your call. Each one is genuinely zero-stakes: no sign-up decisions to agonise over beyond a free account at claude.ai (these all work in ChatGPT too), no settings, nothing that matters if it goes sideways. You cannot break it, and nobody is watching. There’s no test at the end. There’s barely a guide at the end.

The nothing-to-lose list

Five asks. Each comes with the exact words to paste, so the blank box never gets to be scary, and a note on what you’ll notice - because the noticing is where the confidence actually comes from.

1. The thing you’ve always pretended to understand

What it is: everyone has one. Interest rates. The offside rule. What “the cloud” actually is. Inflation. You’ve nodded along for years because the moment to ask passed sometime around 2011.

How to do it:

Nick this one, fill in the blank:

Explain [the thing] to me like I'm a smart adult who has somehow never had it explained properly. Plain language, no jargon - if a technical term is unavoidable, define it in brackets. Keep it short. Then ask me if any bit didn't land, and be ready for follow-up questions - I have years of them saved up.

What you’ll notice: it doesn’t judge. No raised eyebrow, no “wait, you don’t know that?”. You can ask the follow-up you’d never ask a person - “ok but why does the interest rate change anything” - and then the follow-up to that. For a lot of people this is the first moment AI feels less like technology and more like relief.

2. Dinner from your actual fridge

What it is: not a recipe search. A “here is my specific fridge, what can I actually do” conversation.

How to do it: go and look in the fridge. Genuinely - go and look. Then:

Nick this one:

Here's what's in my fridge and cupboards: [list it roughly - "half a bag of spinach, eggs, some slightly tired peppers, rice, a heel of cheddar" is perfect]. Give me three dinner ideas using mostly this, each under 30 minutes, and tell me the one small thing worth buying if I fancy an upgrade.

What you’ll notice: it works with your reality. The slightly tired peppers are fine. Nothing needs to be impressive or complete, and it never says “first, source some sumac”. Most tools make you feel like your life should be tidier before you’re allowed to use them… this one meets you at the half-bag of spinach.

3. The confusing letter

What it is: the one on the side, or buried in your inbox. From the council, the bank, HMRC, the pension people, the school. The one you’ve read twice and still couldn’t say what it wants.

How to do it:

Nick this one:

I'm going to paste in a letter I find confusing. In two sentences, tell me what they're actually saying and whether I need to do anything. If I do, tell me what and by when. Flag anything I should double-check with the sender. Here it is: [paste the letter]

One quiet rule while you’re at it: cover or leave out account numbers and reference numbers before pasting. The letter’s meaning survives without them, and it’s a good habit from day one.

What you’ll notice: the exhale. Three paragraphs of officialese usually boil down to “they’ve changed a date, you don’t need to do anything”, and hearing that in plain words takes about four seconds. Many people tell me this was the ask that converted them - not the clever stuff, this.

4. Three ideas for a present

What it is: there’s always someone. The birthday’s in three weeks and your current plan is “panic near the shops”.

How to do it: the trick is describing the person like you’d describe them to a friend, not a form:

Nick this one, fill in the blanks:

I need a present for [who they are to you], budget around [amount]. About them: [age-ish, what they're into, one true detail]. Things that have landed well before: [anything]. Things to avoid: [anything]. Give me three ideas: one safe, one thoughtful, one a bit leftfield.

To show you what a good fill-in looks like: “my father-in-law, about £30, early 70s, into birdwatching and complaining about the council, loved the flat cap we got him, avoid anything that needs a smartphone”. That one true detail - the complaining, the flat cap - is what turns generic ideas into “oh, he’d actually love that”.

What you’ll notice: you can haggle. Say “cheaper”, “less naff”, “more like the second one but for the garden” and it just… tries again, cheerfully, as many times as you like. No shop assistant to feel guilty about. The first answer is a first draft, and nudging it is the whole skill.

5. Ask it to plan your Saturday

What it is: the softest possible taste of AI-as-assistant. Not your career, not your finances. One Saturday.

How to do it: on Friday evening or Saturday morning:

Nick this one, fill in the blanks:

Plan my Saturday. Things that need to happen: [the fixed stuff - "big shop, kids' swimming at 11, ring my sister"]. Things I'd like to happen: [the hopes - "an hour in the garden, actually sit down at some point"]. My energy level is [honest number out of 10]. Build me a rough, forgiving plan with proper gaps in it, and tell me what to drop first if the day goes wobbly.

What you’ll notice: two things. First, it balances - it takes “actually sit down at some point” as seriously as the swimming lesson, which is more than most of us do for ourselves. Second, and this is the important one: you’re allowed to ignore the plan. Follow it for an hour, abandon it at lunch, doesn’t matter. It’s a suggestion from something with no feelings to hurt. Noticing that you’re in charge here is quietly the point of the whole week.

Honesty corner

Small honest note, even in the gentle guide. AI sometimes gets facts wrong while sounding sure of itself, so anything that matters - a deadline in that letter, a claim about your money - deserves a double-check against the original. It can sound a bit formal until you tell it not to (“again, but like a friend would say it” works a treat). And on the free plan you might occasionally be asked to come back in a few hours - that’s a usage limit resetting, not you being banned, and honestly it means you were properly using the thing.

If you do nothing else

The letter. Ask number three. It takes the least effort, gives the most relief, and it’s the one whose usefulness nobody argues with afterwards.

And that’s the week. Five small asks, nothing broken, nobody watching - and somewhere in there, probably the letter, possibly the fridge, a first moment of genuine surprise: it did in seconds something that usually costs you real effort. That moment is the point. Everyone who’s “good at AI” started exactly there.

Whenever you feel ready for the proper setup - a free account that remembers you and answers the way you like - Your First 30 Minutes With Claude AI will walk you through it with a cuppa. No rush at all. It’ll still be there next week. 🌱