I’d been asking AI questions for months — the same way I used to ask Google. A question, an answer, a closed tab. Useful, but nothing more than that.

Then one day Copilot put a suggestion in front of me: Try making flash cards for the 50 US states. Not a question to ask. A thing to build. I felt curious. So in March 2026, sitting at my kitchen table, I decided to actually try it — make something, not just ask about something.

Two hours later I was still there.

Fifty states. Fifty capitals. Copilot kept submitting cards to me one at a time, asking me to imagine — to actually picture — an image for the capital and the state, over and over, the same request dressed up slightly differently each time. It wasn’t fast and it wasn’t simple, but it was working, in the slow, grinding way these things sometimes work, and I kept going because I was nearly there.

Then it asked me to describe a flash card for Indianapolis, Hawaii.

Indianapolis is not in Hawaii. Indianapolis has never been in Hawaii. Somewhere in two hours of talking, Copilot had lost the plot entirely, and it asked me, in complete sincerity, to help it build a card for a state capital that does not exist.

That’s when I cried. Not because a chatbot made an odd mistake — that happens, that’s forgivable on its own. It was the two hours underneath it. I’d wanted this one small thing to work, I’d put in the time, and it had unravelled into nonsense right near the end, and I didn’t understand why.

A short pause, because the why actually matters.

What a context window is, and why it ate my flash cards

Here’s the thing nobody explained to me at the time, so let me explain it to you. Every AI conversation has something called a context window — think of it as a conveyor belt. Everything you’ve said, and everything it’s said back, gets placed on that belt at the start of the conversation. The belt only holds so much. As the conversation keeps going, new things get added to the front, and the oldest things eventually fall off the back end and are gone.

Copilot is extremely verbose. It doesn’t just answer — it explains itself, expands, elaborates, fills the belt with its own commentary as much as with the actual task. Fifty rounds of “imagine an image for this capital and this state” adds up fast. Which means the belt fills up fast, and the early instructions — the actual point of the conversation — fall off the back long before the conversation feels finished. By the time it asked about Indianapolis, Hawaii, the part of the belt that remembered we were talking about real state capitals had already gone.

That’s not a quirk. That’s the mechanism. And once I understood it, the two hours — and the tears at the end of them — made a different kind of sense.

What happened after

I haven’t opened that Copilot conversation since. It’s sitting there, untouched, since March — and if I opened it today, its own memory of how it started would be long gone, fallen off the back of its own conveyor belt months ago.

But that afternoon did something. Still at the same laptop, I opened Claude.ai in the browser for the first time. I’d talked to Claude on my phone before, in passing, the way you talk to anything new and not yet trusted. This was different.

I described what I wanted. We talked it through. And in under an hour, I had it — the same idea Copilot had spent two hours grinding me through and still got wrong at the finish line, built properly, with no wasted rounds and no Indianapolis in Hawaii.

That’s the part I actually want you to sit with. Not the tears. The hours. I didn’t write code. I didn’t follow a tutorial. I had a conversation — just a conversation — and a working thing came out the other end. Two hours of one tool fighting itself versus one hour of just talking to another. That contrast is the whole post.

An invitation, not a warning

I’ve told you about the tears because they’re true, and because I think the cleaned-up version of this story — the one where I sound capable from the first paragraph — wouldn’t actually help anyone. But the tears aren’t the headline. The hour is the headline.

If you’ve been circling AI the way I circled it for months — curious, a bit wary, not quite sure what you’d even use it for — I’d say: go and tinker. Not with a plan. Not with a course you’ve enrolled in. Just open something and try to build a small, slightly silly thing. A flash card set. A list. A tool for the one tiny, specific problem that’s been bothering you for years and never seemed worth fixing properly.

You might hit your own Indianapolis, Hawaii. If you do, now you know why, and you know it passes.

But here’s what I actually want you to take from this: that Tuesday afternoon, the one where I opened Claude in a browser for the first time, a conversation — nothing more than that — turned into a thing I’d actually built, in less time than I’d just lost going nowhere. That was it. That was the whole moment.

It doesn’t announce itself. It just leaves you holding something that works, made out of nothing but talking, and you realise, a beat too late, that you’re not the same person who opened the laptop an hour ago.

Have the courage to tinker. With any luck, no two-hour detours required. But even if there are — they’re not the end of the story. They’re just the part before the good bit.

I keep coming back to a question about the Manager Who Built the Tools, and it’s this: why did my bosses never give me the specific tools I actually needed?

They gave me plenty. The daily balance program. The stocktake system. The software that ran the service for handling people’s affairs while they were away — the routine stuff, the kind of thing every office in the organisation needed, so somebody built it centrally and handed it down.

But nobody ever gave me a tool for the actual texture of my job. Writing letters to recalcitrant payers. Telling someone we could, or couldn’t, resolve their complaint. Asking a person to come into the office in person, in writing, in the right tone. Ten letters a day, sometimes — different recipient, same letter, typed out from scratch every single time. Because we didn’t have a photocopier either, not at first.

This was the pre-PC era. I didn’t stay in it long. The moment a PC landed on a desk near me, I was hooked. Something new — and more importantly, something I could actually point at my own problem.

So I got stuck in. I started building the small solutions for the small jobs nobody upstairs had thought worth solving — the ten-letters-a-day problem, the same-paragraph-different-name problem. Nobody asked me to. Nobody trained me to. I just couldn’t stand typing the same sentence out by hand one more time when there was a machine sitting right there that clearly could do better.

I think if I’d stayed too long in a non-tech environment, I’d have ended up some kind of inventor anyway — looking for ways to improve the daily grind because that’s just how my brain works. Lucky for me, the company actually put money where its mouth was. PCs arrived. Dot-matrix printers. Eventually photocopiers. And then the telex machine.

I remember the thrill of that one specifically — typing a message at my desk out the back of Oakleigh and sending it, in text, to Karen over at Moorabbin. That felt like magic. Two offices, talking in typed words, instantly. Thirty-five years ago.

Here’s the thing that’s been sitting with me since I started writing about that Manager (past me), and Yol (my Claude AI Assistant), and all the systems I built over four decades: every generation gets handed new tools, and every generation has someone in it who can’t wait to get their hands on them and start pointing them at the boring, repetitive, nobody-thought-to-fix-it parts of the job. The telex was somebody’s AI moment. The photocopier put an end to ten identical letters typed by hand, the same way Claude now puts an end to a whole afternoon of work that used to need a developer and a budget line.

The technology changes completely. The instinct that reaches for it doesn’t change at all.

I was that person with the telex machine. I’m that person with Claude. In between, I was that person with Access, with Excel macros, with whatever the latest thing was that let me stop doing something by hand that a machine could clearly do better.

If you’re reading this and thinking AI is somehow different — too big a leap, too strange a tool, not for people like you — I’d ask you to think about what your own version of the telex machine was. Somewhere in your working life, something new arrived that changed how a small, tedious part of your job got done, and you adapted to it, probably without thinking of yourself as a tech person at all. You’ve done this before. You just didn’t call it that at the time.

This is the same thing. It’s just faster, and it talks back.

There’s a pattern that ran quietly through my career, and it wasn’t the heroic version I sometimes imply. I didn’t build everything. Management introduced plenty of tools over the years — Word, PowerPoint, Excel, SharePoint, ServiceNow, MS Project, Access, Copilot, and before all of that, Symphony, WordPerfect, Quattro Pro, dBASE, Lotus 1-2-3. Some ancient IT-made tools from the ‘80s.

The truth is simpler, quieter, and more accurate:

When something essential didn’t exist — and the work couldn’t move without it — I built the missing piece.

Not because I wanted ownership.
Because the job needed doing.

I feel my generation inherited the baby-boomer’s attitude toward getting things done. You just did it. If the thing didn’t exist, you invented it. It’s a thread that ran through my whole life.

I’m telling you this now, before the stories, because it’s the whole point: AI didn’t hand me a new skill. It handed the same old instinct a much faster way to act on itself.

The first tool I built because nobody else was going to

In 1998, I needed a database. A real one.

Something to store the guts of the work we were doing — the client answers, the logic, the decisions, the specifications.

The company didn’t have one.
Nobody was planning to build one.
Nobody even understood why we needed it.
So I opened MS Access and built AegeanC.

It wasn’t glamorous.
It wasn’t strategic.
It was simply necessary.
And people used it.
Quietly.
Every day.
The tool that grew while I was away

A few years later, while I was off on another project, a colleague built BluC — inspired by AegeanC, extending it, improving it. When I returned, we integrated the two. Over the years, others tinkered with them. Someone built a front-end database. Someone else added automation around 2014.

It was never “my system.”
It was a living thing — shaped by many hands, across many years.
But the bones were familiar.
The logic was familiar.
The purpose was familiar.
And that mattered.

The moment everything changed: MedC.

By 2019, the IT team who used the database daily were exhausted by its slowness. One brilliant man reached out to a colleague in WA — a wizard with Oracle APEX — and together they rebuilt the entire thing into a single, elegant system.

That system became MedC in 2020.

It was fast.
It was modern.
It was everything the old databases had been trying to be.

And even though my career had taken me elsewhere again, I stayed connected to the team. When my project had quiet moments, I’d jump in and help improve MedC— working closely with our developer in WA.

His name was Ulysses, but with the accent, it sounded like “Yol.”

So Yol he became.
And Yol and I — we were unstoppable.
Not because we were special.
Because we were aligned.
I made the ideas.
He built them.
We iterated.
We refined.

We solved real problems for five local teams and five interstate teams who relied on MedC every day.

It was the best kind of collaboration — the kind where the work gets better because two people care enough to keep going.

The year of building before I left

When I learned in April that my tenure would likely end on September 30th, something in me clicked into gear.

Not panic.
Not grief.
Just clarity.

If I was leaving, the knowledge couldn’t leave with me.

So I got to work.

I defined major changes to the way MedC functioned — not for glory, but so the team would have what they needed long after I was gone. I wrote paper after paper. I built a knowledge library. I linked everything: Confluence articles, SharePoint documents, ServiceNow windows, every morsel of process and logic that still lived in my head.

I wasn’t building tools anymore. I was building continuity.

And Yol — patient, brilliant Yol — built everything I threw at him.

When I left in December, MedC was stronger, clearer, and more future-proof than it had ever been.

Not because of me alone. Because of the team. Because of the collaboration. Because of the years of shared effort.

The pattern I didn’t see until much later

So when my ex-staff member said, “AI hasn’t reached us yet,” I felt something old stir.

Not ego.
Not superiority.
Just recognition.

The same instinct that made me build AegeanC in 1998.
The same instinct that made me integrate BluC.
The same instinct that made me work with Yol on MedC.
The same instinct that made me write the knowledge library.

If the work needs something and it doesn’t exist yet, I build the missing piece.

Not because I’m the one who can.

But because I’ve spent a lifetime stepping into the gap between “what we have” and “what we need.”

And AI hasn’t reached them yet for the same reason none of those earlier tools arrived on their own, either. Someone has to walk through the gap first. It was never going to be the org chart. It was always going to be whoever was standing closest to the problem with the nerve to start.

The point of this story

This isn’t a story about being the hero.

It’s a story about being the person who notices the gap.

For me, the gap used to take months to close. A budget request. A developer in WA. A year of writing things down before I walked out the door in December.

Now the gap is an afternoon.

That’s not a small difference. That’s the whole shift this blog is trying to name — the same instinct I had in 1998, except now it doesn’t need Yol, or Access, or permission from anyone at all. It just needs me, a question, and the willingness to start before I feel ready.

I built AegeanC because nobody else was going to.

I’m still that person. The tools just finally caught up to the instinct.

20th December 2024. That’s the date, if you want it precisely. The redundancy was generous — generous enough that saying yes was the only sensible answer — and I said yes, and then I had to work out who I was without the thing that had defined me for forty‑one and a half years.

Nobody warns you about that part.

They tell you about the holidays. The freedom. The lie‑in on a Tuesday. They tell you about the novelty of not having to be anywhere, the luxury of time suddenly unstructured. They don’t tell you that you might spend decades being the person everyone calls, and then one day the calls simply stop — not because you got worse at the job, but because you left the building.

There’s a silence that follows competence.

A silence that feels like a compliment at first — you’ve earned your rest — until it starts to feel like an erasure.

Who are you when nobody needs you?

Who are you when the room you used to hold together no longer exists?

Retirement is sold as an ending.

Nobody mentions the identity vacuum that comes after.

The reinvention

I did what you do. I bought a house in the country. I pulled my parents’ furniture out of storage and set the place up properly, for a dear friend to rent. There were holidays there — friends, family, the good kind of busy. The kind of busy that feels like a life being lived, not a life being managed.

And then winter came down over Melbourne the way it does: bone‑chillingly cold, and the kind of quiet that makes you reach for something to do with your hands.

I scrapbooked.

Eighteen albums, that first winter. Perseverance was my middle name, and I mean that as a compliment to myself, because eighteen albums is not nothing. It’s a small act of devotion, repeated until it becomes a rhythm. The final album from that winter is still sitting on my dining table right now. It’s not a dead project. I’ll never tire of scrapbooking.

But somewhere in those wintry weeks, something else happened.

Something subtle.

Something I didn’t have language for at the time.

The particular exhaustion of forty‑one and a half years of being indispensable — the kind you don’t notice you’re carrying until it starts to lift — began to lift. And underneath it was a restlessness I hadn’t expected.

A capable person with no problem left to solve is not actually at peace.

She’s just quiet.
Quiet is not the same as fulfilled.
Quiet is not the same as done.
Quiet is just the absence of noise — not the presence of purpose.

The thing nobody tells you about competence

I didn’t want the cold, quiet version of winter to be my only answer for 2026. I didn’t want to become someone who filled her days with tasks that kept her hands busy but her mind under‑used. I didn’t want to be a woman whose best thinking was behind her.

So when AI arrived in my life — properly arrived, not the curious poke I’d given it before — it didn’t find a woman who needed to learn how to think.

It found a woman who’d been doing nothing else but, for four decades, with nowhere to focus it now.

A few very deep conversations about a medical history I needed answered properly.
Research into what makes a forever home actually a forever home.
Then I built a memory tool. A game. A planner that worked the way my brain actually works, instead of the way every productivity app assumes everyone’s brain works. A database for keeping track. A place to store all my stories, to start writing my book.

And somewhere in the building of those things, I came back online.

Ridiculously productive again.

The smartest person in the room — except the room was empty, and that was, for a little while, the funniest and saddest sentence I could have written about myself.

But here’s the thing about empty rooms: they don’t stay empty if you start building things inside them.

People wander in.

Ideas wander in.

Purpose wanders in.

And suddenly you’re not retired.

You’re repurposed.

A different vessel, not a different boat

Retirement is sold to you as the end of useful work.

Here’s what nobody mentions: the skills don’t retire with you.

The problem‑solving, the translating, the relentless instinct to find out and fix and figure it out — all of that was just sitting there, fully intact, waiting for somewhere to go.

Competence doesn’t evaporate.

It waits.

And when AI arrived, it didn’t give me a new identity.

It gave me a new vessel for the identity I already had.

I didn’t miss the boat.

I just hadn’t found the right one yet.

This is the first post in a story I’m going to tell you properly, over the next few weeks — everything I’ve built since that winter lifted, and what it’s actually meant to come back alive.

There’s a fair bit to tell.

Stay with me.

It came through on a WhatsApp group chat, the way these things do now.
One of my ex‑staff, mid‑conversation, said the line I’ve heard a dozen versions of:

“AI hasn’t reached us yet.”

Not in Operations, anyway. They knew it was happening somewhere — the shop floor, maybe, or the Call Centre, definitely Sales and Marketing — but it hadn’t arrived where they sat. They were waiting for it to.

And the moment I read it, something in me tightened.
A sense of déjà vu.
A flicker of irritation.
Not at him — never at him — but at the fact that the team who could benefit the most from AI were the ones who’d been left out of the conversation entirely.

And a bigger irritation, the kind that sits low in the chest:
they could help themselves.
They could pick up their phones and start using it today.
They didn’t need a rollout.
They didn’t need a licence.
They didn’t need permission.

But they were waiting for it anyway.

And in the recesses of my brain, a tiny idea stirred:
Maybe there’s an opportunity here. Maybe I can help them.

The wait that used to make sense

There was a time when waiting was the right call.

In our workplace — when I was still there — my team lived twelve weeks ahead of the calendar. We were working on October’s output when it was actually July. It had a strange effect on all of us. Disorienting. Dislocating. Like living in a timezone nobody else recognised.

And for me, it was worse.
Because at the time, I was addicted to Garden Answer — a gardener in Ontario, Oregon USA, appearing on my TV every night, in a climate six months out of sync with mine. So I was living twelve weeks ahead at work and six months behind at home. I didn’t know what day, week, month, or season I was in. We had to recenter our internal calendars constantly. In July, it felt like Christmas was three months away.

But here’s the real point:
we weren’t the ones waiting.
Everyone else was waiting on us.

Waiting for us to write the specs.
Waiting for us to finish the analysis.
Waiting for us to hand over the documents so the developers could build the software.

The waiting was crushing.
The waiting was structural.
The waiting made sense.

And that instinct — the one that says “I can’t start until someone gives me the tool” — is still running.

It’s just running in a world that changed underneath it.

What I did that same afternoon

I didn’t say any of this in the WhatsApp chat.

What I did instead was move from my desk — where the sunlight was streaming in so brightly behind me that it washed out my laptop screen — to the couch, where I could actually see what I was doing.

I’d spent the last two days churning out three apps and feeling quietly impressed with myself.
I’m pretty good at this, I thought.
I wonder what I can remember of the work I used to do 18 months ago.

I hadn’t thought about Business Analysis in ages.
Once, in a team chat, someone used an acronym for the database I’d invented in 1998 — the name of the database I’d built — and I couldn’t even remember what it stood for. That’s how far away that world felt.

But the idea had always been within me:
a tool that could quiz our clients and turn their answers into a specification.
The guts of the job.
The 80% of the day that nobody enjoyed.

I’d wanted the company to fund it for years.
They never did.

So I opened a blank screen.
I fed Claude the questions I remembered, the structure of the spec, the logic of the workflow.
And in an hour — an hour — it built the tool.

I was agog.

Astonished at how much I still remembered.
Astonished at how quickly it came together.
Astonished that the thing I’d wanted budget for — the thing that would have saved my team hundreds of hours — was suddenly sitting in front of me, fully functional.

A few refinements and it could be perfect for them.
My ex‑staff.
My team.

Imagine if I were still there.

And then the uncomfortable thought arrived — the one I’ll write about properly another time:
if 80% of their day could be automated, that would put them in the crosshairs for the next round of redundancies.

I couldn’t do that to them.
Not yet.
Not like that.

The objection that’s actually fair

I can hear the reply already, because I know the people who’d give it:

“We can’t install anything on our work computers.”
“We can’t send anything off the premises.”
“Everything’s locked down.”

And they’re right.
That’s not an excuse dressed up as a barrier.
That’s a real barrier.

Their IT environment is locked down so tightly that nothing gets in or out without Big Brother seeing it. No software can be installed. All emails are scanned. It’s a fortress.

And I know exactly what would happen if I handed them an AI tool:
they’d reject it immediately.
Not because they didn’t want it.
But because they’d assume the fence was impenetrable.

So let me be precise about what I’m actually saying.

The permission you do have

If you’re in that environment, the tool you’ve been given is, maybe Copilot. If you’re very lucky, you’ve got ChatGPT or Claude as your company’s pre-installed LLM.
And the honest problem isn’t that AI hasn’t reached you.

It’s that you’ve been handed something and nobody taught you how to use it well.

Copilot loses context fast.
It rambles unless you manage it.
It gets confused if you stack five tasks into one message.

But if you give it:

• short, specific instructions

• one task at a time

• a fresh conversation when it starts circling

…it becomes useful.
Not magical.
Not perfect.
But useful.

It’s a smaller permission than “go build whatever you want.”
But it’s real.
And it’s available to you today.

For everyone else — the people with a phone in their pocket and no IT department watching what they install — the bigger permission really is sitting there unused.

But if you’re the person who genuinely can’t get past the locked‑down environment:
the fence is real.

What’s also real is that you’ve still got more room inside it than you’ve been using.

The worker I’m writing to

I’m writing to the person who’s brilliant at their job but cautious.
The one who waits for the calendar invite.
The one who assumes someone else will tell them when it’s time to begin.
The one who has spent a career being responsible, reliable, and respectful of process.

The one who doesn’t realise the process changed without telling them.

The moment I stopped waiting

His message — “AI hasn’t reached us yet” — made me want to be back at work for a moment.
Not for the job.
For the team.

To implement something.
To enforce it.
To give them the tool they’d never go looking for.

Because my whole career worked like this:
if I didn’t build it, it didn’t exist.

The systems, the templates, the databases, the procedures — they all had my name in the Author column. I invented the tools I needed because nobody ever handed them to me.

And that’s when I realised:
I stopped waiting years ago.
Not because I was brave.
But because I had no choice.

The permission was never going to come from above.
It had to come from me.

And now — in this new world — it has to come from you.

He’s on the internet every day. Buy-swap-sell groups, the local noticeboard, the niche community for the exact thing he collects, the Marketplace deal that saved him four hundred dollars on a trailer he’d been hunting for weeks. He moves through it with the easy confidence of someone who actually knows what he’s doing — because he does. Nobody handed him that. He worked it out himself, one slightly annoying afternoon at a time, until annoying turned into automatic.

So when people talk about “the internet,” he nods along. He’s not being polite. He genuinely believes he’s across it. He’s on it constantly. He’s good at it. As far as he’s concerned, he’s mastered the thing.

The one room he knows extremely well

Here’s what I haven’t told him yet, because I think it needs saying gently: Facebook is one room.

A very good room, to be fair. Well-decorated. Heavily populated. Algorithmically optimised within an inch of its life to keep him exactly where he is, scrolling, buying, selling, finding his people. He has every reason to feel competent in it. He is competent in it.

But it’s one room in a building he doesn’t know exists.

The rest of the building

I’m not saying this to make him feel small. He’d hate that, and frankly he wouldn’t deserve it. He’s not avoiding the rest of the building out of fear, or stubbornness, or some principled stand against the broligarchy this time. He genuinely doesn’t know it’s there. Nobody ever told him there was more than the room he’d already mastered.

The building has a lot more in it than buy-swap-sell. A conversation that can help him think through a decision the way a smart friend would, at midnight, without judging him for asking the same question three different ways before he gets the words right. A tool that can take the mess in his head and turn it into something organised, the way he organises a Marketplace search by suburb and price and condition. Something that rewards exactly the instincts he’s already built — patience, persistence, the willingness to try something five ways before giving up on it.

He thinks those instincts only work on Facebook. They don’t. They’re the instincts.

Why this door is easier than the last one

The thing that strikes me, watching him move through Marketplace with total confidence, is that he already did the hard part. He already learned to navigate something complicated, with no instructions, by trial and error, until it made sense. That’s the entire skill required to get good at a conversation with AI. He’s not starting from zero. He’s starting from years of practice he doesn’t recognise as practice, because he never called it that.

I’m not asking him to become someone new. I’m asking him to walk down the hall.

He found the trailer. He found the rare thing nobody else could track down. He found the deal everyone else missed. The room he’s already mastered taught him exactly the skills he needs for the next one.

He just has to open the door.

He once drove to Darwin to buy a car.

Not because there wasn’t a perfectly good car closer to home. Because he’d found one on Facebook Marketplace that was a genuine bargain, and he is the kind of person who will drive the length of the country to claim a win. He knew which groups to search. He knew how to spot a real listing from a scam. He knew how to message a stranger in a way that got a fast, friendly reply instead of silence.

That’s not luck. That’s a skill. He just doesn’t call it one.

The man who isn’t a tech person

Ask him about AI and you’ll get the full performance. I wouldn’t know what to do. Too complicated. AI is the devil. And the one that actually explains the rest of them: I’m not paying the broligarchy.

He has opinions about Silicon Valley billionaires that he will share, unprompted, at len

gth. Some of them are fair. Most of them are doing a lot of work to avoid saying the real thing, which is: I don’t think I’m the kind of person who could do this.

Meanwhile he has spent years getting genuinely good at something most people his age never bothered to learn. Facebook groups. Marketplace. Finding the right community for the right question. Reading a thread fast enough to know who’s trustworthy and who’s wasting his time. Nobody taught him any of it. No course, no tutorial, no younger relative sitting him down. Just time, persistence, and the particular satisfaction of beating a platform that defeats plenty of people his age.

He built that skill the exact same way anyone builds a skill with AI. Trial. Error. Annoyance. Doing it again slightly differently. Eventually, mastery — quiet, earned, never once described to himself as “technical.”

The skills transfer completely

Here’s what he’s actually been doing every time he opens that app:

Reading how information moves through a network of strangers. Understanding what people actually need versus what they say they need. Finding the signal in a noisy feed. Navigating a complicated system by poking at it until it gives up its logic.

That is not a small set of skills. That is, more or less, the entire skill set required to get something useful out of an AI conversation. Knowing what to ask. Reading what comes back. Adjusting when the first attempt misses. Persisting past the part where it would be easier to give up and say it’s too complicated.

He has been doing the hard part for years. He just did it on a platform that doesn’t ask him to pay anyone he disapproves of.

The ceiling that was never there

I’m not going to tell him AI isn’t run by the people he thinks it’s run by. He’s not wrong about that, and it’s not really the argument that’s going to move him.

But I will say this: the ceiling he’s been respecting — I’m not technical enough — was never a ceiling. It was just the last thing he happened to try before he found something he was good at. He hit the same wall with Marketplace once. He just didn’t stop at it.

He drove to Darwin for a car. The man who does that is not short on capability. He’s just decided, for reasons that have very little to do with capability, that this particular door isn’t his to walk through.

I think he’s wrong about that. I think he’d be terrifyingly good at it.

My mother is gone now.

She died in 2014, at the end of a long leaving — the kind that dementia makes, slow and then faster and then in ways that stopped surprising us because we had run out of the capacity to be surprised.

But for the years before that ending, she was still here. Physically present — breathing, sleeping, sometimes smiling at something none of us could see. The body that held her was still in the world. Her hands were the same hands. Her face, in certain lights, at certain angles, was the face I had known my whole life.

The person who lived behind that face — the mother who knew my name, who had opinions and memories and the particular way she held her cup — she had been leaving for years before she left entirely.

The grief of dementia is that it arrives before the loss. You mourn someone who is still in the room. You carry the weight of an ending that hasn’t ended yet, that keeps not ending, that stretches out in front of you with no clear horizon and no moment you are permitted to say: now. Now I can grieve.

There was no funeral yet. There was just the Tuesday. And the Wednesday. And the Thursday after that.

The people in the story

My father was the carer.

He watched his wife become, as he sometimes said with a gentleness that broke my heart, a child again. He managed her days with a patience and a love that I could not have predicted and cannot adequately describe. He was exhausted in the way that people who have been exhausted for so long that it has become their baseline are exhausted — not dramatically, not visibly, just always. Underneath everything. The tiredness that didn’t lift because there was no morning after it.

My sister was twenty-five kilometres away. I was in another state.

We were all trying to help in our own ways, from our own distances, with our own versions of the grief that we couldn’t quite name to each other — because naming it felt like a betrayal of the people still living it. Dad was suffering enough. We didn’t want to add our grief to his. We didn’t want to make him responsible for holding us up when he was already holding everything else.

So we held it separately.

Which meant we held it alone.

The thing about turning to each other

We couldn’t.

That’s the honest answer, and I want to say it plainly because I think it’s true for more families than will admit it.

We loved each other. We were not estranged or difficult or unable to communicate. We were just — each carrying something that felt too heavy to hand to someone who was already carrying their own version of the same thing.

Dad’s grief was watching his wife. My sister’s grief was proximity — the Tuesday visits, the face across the table, the hands that were the same hands. My grief was distance — the not being there, the phone calls that ended and left me alone with what had been said.

Three different griefs. The same loss. And none of us quite knowing how to support each other without adding to the weight.

At 3am, when the pain of knowing and not being able to fix became too much to hold — there was nobody to call. Dad was asleep, or awake and shouldn’t be burdened. My sister was asleep, or awake with her own version of the same thing. The friends who loved me didn’t quite understand, because you cannot understand this from the outside, not really, not until you have sat with it.

I held it alone.

What I didn’t have

AI didn’t exist in the way it exists now. Not the kind that could sit with me at 3am and receive what I was carrying without needing to be protected from it.

I want to tell you what I would have done with it, because I know exactly what I needed and didn’t have.

I needed somewhere to put the question I couldn’t ask Dad because it would hurt him. The thing I was trying to understand about her medication, her behaviour, the change I’d noticed on the last visit that I didn’t know how to interpret. The legal question that arrived without warning — about power of attorney, about what happens when certain decisions need to be made, about what her wishes had been before she lost the ability to express them.

Those questions had answers. Not easy answers, not complete answers, but answers — information, framework, language, the shape of what we were dealing with so that the fear had something to hold onto besides itself.

And then there was the other kind of thing I needed. Not a question. Just somewhere to say: I don’t know how to do this. I don’t know how to grieve someone who is still here. I don’t know how to support my father without breaking down in front of him. I don’t know how to be far away and useful at the same time.

There was nowhere to say that. Not without burdening someone who was already burdened.

If AI had existed the way it exists now — if I could have opened a conversation at 3am and said the thing I was actually feeling, not the managed version, not the one edited for the people I loved — I would have used it every single night.

The invisible carer

I want to name my father properly in this post, even without naming him.

He had no official support. No training, no respite, no system that saw him clearly as someone who also needed care. He was the carer — which meant he was the one the system pointed everything at, and also the one the system most consistently forgot to ask: and how are you?

AI cannot give the carer what they actually need. It cannot give them rest, or company, or the presence of the people who love them. It cannot give them their person back.

But it can answer the question they are too tired to google at 11pm. It can help them understand the behaviour that frightened them today. It can help draft the letter to the specialist, navigate the aged care system, find the form that needs to be filled in before the deadline nobody knew existed.

It can be available when everyone else is asleep.

And for the person doing the hardest job in the family for years, without a title or a salary or a day off — available, at any hour, without needing anything in return — is worth more than I can easily say.

What I know now

My mother died in 2014. My father followed her, a few years later — the way some people do, when the person they were built around is gone.

The story is complete now. I know how it ended.

What I also know, from the other side of it, is what I was carrying without the tools that exist now. The 3am questions that had no home. The grief that had nowhere to go that didn’t cost someone else something. The fear that had no shape because I didn’t have the language to give it one.

If you are in it right now — if your person is still here, partly, intermittently, in moments that arrive without warning and disappear just as fast — I want you to know that the tool exists.

It won’t fix it. Nothing fixes it.

But it will sit with you at 3am. It will receive what you’re carrying without flinching. It will answer the question you can’t ask anyone else, and hold the weight of the thing you can’t say out loud to the people you love.

I did this without it.

You don’t have to.

He didn’t know what was happening to him.

That’s the part I keep coming back to, when I think about that night. He was there — physically present, in the bed, with the nurses moving around him with the particular purposeful speed that you learn, eventually, to read as urgency. But he wasn’t tracking. The fever had taken something from him temporarily — the ability to follow, to understand, to ask the right question at the right moment.

So I asked them.

Not because I knew what to ask. Because someone had to, and I was the one standing up.

What I didn’t know

I want to be honest about where I started that night, because the ending only makes sense from the beginning.

I didn’t know what Streptococcal Pneumonia was. I knew the word pneumonia — enough to know it was serious, not enough to understand why the nurses in the Emergency Department were moving with that particular speed, why the first available bed was being found with such urgency, why the temperature and heart rate on the monitor were producing expressions I couldn’t quite read.

I knew something was very wrong. I didn’t know the shape of it.

So I did what I do. I found out.

I typed what I knew into Claude — the diagnosis they’d mentioned, the temperature, the speed of the nurses, the words I’d caught and half-understood — and I started reading. Not panicking. Reading. Because facts have always been how I stay upright when everything else is moving too fast.

What came back was a small, precise education delivered at exactly the moment I needed it.

_{Note: The patient's face in this photograph has been digitally altered using AI to protect their identity and medical privacy. The surrounding environment and medical equipment remain an accurate representation of the event.}

Streptococcal Pneumonia. The temperature and what it meant. Why the Infectious Diseases Unit would become involved. What empyema was — fluid between the layers of the pleura, the membrane surrounding the lungs — and why that particular complication produced the particular expressions on the faces of the people moving around him.

I learned the words. And then I learned what the words meant. And then I understood, with a clarity that was frightening and stabilising in equal measure, exactly how close we were to something I didn’t want to name.

The knowing nod

I have nurses in my life. More than I realised, until I started telling people what had happened.

When I told them — when I said empyema, when I mentioned the pleural involvement, when I used the language I’d borrowed from an hour of reading in an emergency waiting room — something happened on their faces.

The knowing nod.

Not surprise. Recognition. The nod of someone who understands, professionally and viscerally, exactly what those words mean in combination. Who knows the trajectory. Who has seen it. Who is looking at me with a kind of quiet respect for the fact that I know what I’m talking about — and a kind of quiet relief that I do, because it means I understand what we were dealing with.

I cannot tell you what that nod meant to me.

Not because it made me feel clever. Because it confirmed that what I had understood was real. That the fear had been proportionate. That I hadn’t been catastrophising in a waiting room at midnight — I had been accurately reading a situation that warranted exactly the level of attention I’d given it.

The facts had held me up. And now they were holding up everyone else who loved him too.

The other job

Here’s the thing about being the person who knows what’s happening in a medical crisis.

You become the bridge.

Not just to the patient — to everyone. The family members who call and ask and need to understand. The friends who want to help but don’t know what they’re helping with. The people who love him and are frightened and need someone to say, clearly and calmly, here is what is happening, here is what it means, here is what comes next.

That job has no official title. There is no lanyard for it, no role description, no training. You are simply the person who is standing up when everyone else needs someone to stand up — and you are doing it while also being frightened, also being exhausted, also being the person who loves him and doesn’t want to think too clearly about what the alternative looks like.

I could do that job because I had the facts.

Without the facts — without the hour of reading in the waiting room, without the words that gave the fear a shape I could hold — I would have been as lost as he was. Present but not tracking. Standing up but not steady.

AI didn’t make me a doctor. It didn’t replace the nurses or the Infectious Diseases Unit or the medical team who knew exactly what to do and did it with skill and speed.

It made me the most informed non-medical person in that building. And in that particular role — the unofficial, unlanyarded, absolutely essential role of the person who explains to everyone else — that was exactly what was needed.

The knowledge made me feel strong when I wanted to crumple.

What it means for you

You may never stand in an emergency department at midnight trying to understand words that will tell you how frightened to be.

But you will, at some point, be the person standing up when someone you love can’t. The one who has to understand, and then explain, and then hold everyone else together while also holding yourself.

That moment doesn’t announce itself. It arrives without warning and requires everything you have.

AI won’t give you the courage for it. That part is yours.

But it will give you the facts. And facts, it turns out, are how some of us stay upright when everything else is moving too fast.

She got the call on a Tuesday.

Her sister’s voice was steady — the particular steadiness of someone who has just received news and hasn’t yet decided how to feel about it. A diagnosis. The name of something that was now living in her sister’s body, that had a medical term and a treatment pathway and a prognosis that the doctor had explained in language that had washed over both of them like a tide coming in too fast to read.

She wrote down the word. She couldn’t spell it. She wasn’t sure she’d heard it correctly.

She said: I’m coming over.

And then she sat in her car in the car park of wherever she was when the call came, and she thought: what do I actually do now?

The gap between caring and knowing

Here is the thing nobody tells you about loving someone through a health crisis.

Caring is not the same as knowing. And knowing is what helps.

She would do anything for her sister. That part was simple — the love was not in question, had never been in question, would never be in question. But love, on its own, doesn’t know what empyema means. Love doesn’t know which questions to ask at the oncology appointment, or what the difference is between a Stage 2 and a Stage 3 diagnosis, or why the doctor mentioned a particular specialist in a particular tone of voice that felt significant, but she couldn’t explain why.

Love shows up. But showing up without the knowledge can feel, some days, like not quite enough.

She had grown up in a family without doctors in it. No medical background, no professional frame of reference, no cousin who could be texted at midnight with the question she was too embarrassed to ask the actual medical team. Just her, and her sister, and a diagnosis she couldn’t spell, and the particular helplessness of wanting to do something useful and not knowing what useful looked like.

What she did in the car park

She typed the word into her phone. Her best guess at the spelling — close enough, it turned out, for the AI to understand what she meant.

What came back was not a medical consultation. It was not a diagnosis or a treatment recommendation or anything that replaced the team of professionals who were, at that moment, already doing their jobs with skill and care.

It was a plain language explanation of what the word meant. What the diagnosis involved. What the treatment pathway typically looked like, and what questions were worth asking at the first appointment. What the staging meant. What the specialist the doctor had mentioned did, specifically, and why that referral might have been made.

She sat in the car park for forty minutes.

She came out knowing enough to ask the right questions. Not all of them — there were questions she wouldn’t know to ask for months yet, questions that only arrive after you’ve been living alongside something for long enough to understand its shape. But the first ones. The ones that would tell her sister: I looked it up. I came prepared. I am not going to make you explain everything from the beginning every time we talk, because I already know the beginning.

That is its own form of love.

The family with a doctor in it

I want to name something that doesn’t get named enough.

In a health crisis, information is power. And it has always been unevenly distributed.

The family with a doctor in it — a GP cousin, a nurse sister, a friend from medical school — asks different questions than the family without. They know which results to ask for. They know when the answer they’ve been given is incomplete. They know the difference between a wait-and-see and an act-now, even when both are delivered in the same measured tone.

The family without a doctor in it relies on what they can find, and what they can understand, and how brave they feel about asking questions that might reveal how little they know.

AI doesn’t give her a doctor in the family. It gives her access to the same quality of informed curiosity that used to require knowing the right people. The ability to walk into an appointment with questions that show she has done the work — not to impress anyone, but because her sister deserves an advocate who has done the work.

She still can’t fix it. No amount of research changes the diagnosis or speeds the treatment or removes the fear that arrives at 3am when everything is quiet and the love has nowhere useful to go.

But she can show up informed.

And showing up informed, it turns out, is something her sister can feel. The difference between the visitor who arrives and needs to be managed, and the one who arrives and can be leaned on. The one who asks the question the patient forgot to ask. The one who remembers what the doctor said when the patient was too frightened to retain it.

That’s not a medical skill.

It’s just love, equipped.

On the night of the 14th of April 1912, the RMS Titanic was making excellent time.

The crossing was going well. The ship was magnificent. The destination was New York. The speed felt like confidence — like proof that everything was working exactly as it should.

The iceberg was on the charts. The warnings had come in by wireless. They were noted, and then the ship kept going, because the destination mattered and the schedule mattered and stopping to reconsider felt like weakness when everything looked so good from the bridge.

I think about that night sometimes when I think about a man I used to work with.

The man who had everything

He had everything, by any reasonable measure.

A high-paying, high-powered job in Information Technology. A nuclear family — children nearly grown, the kind of almost-adults you look at and think: we did something right. A house in a good suburb, freshly renovated, worth half a million dollars more than it was five years ago. A career trajectory that pointed in one direction: up.

He was making excellent time.

And he was dying.

Not literally. But something in him was. The pressure of sales targets and quarterly deadlines and the particular grinding weight of working for a large company’s bottom line, year after year, without being able to answer the question that had started arriving in the early hours of the morning, quiet and persistent and impossible to sleep through:

What is this actually for?

Not the job. Not the house. Not the renovated kitchen or the nearly-grown children or the suburb with the good schools. All of that was real and earned and worth protecting.

But underneath it — underneath the trajectory and the targets and the excellent time — there was a man who had been going in one direction for so long that he’d stopped checking whether it was the right one.

You can’t ask the band

Here’s the thing about a mid-life crisis that nobody tells you.

Not really. Not to the people who know you. Because the people who know you are also invested in the version of you that has everything — the version that is making excellent time and doesn’t need to reconsider. To say out loud, to someone who loves you, that you don’t know if any of this is right anymore — that’s a conversation that has consequences. For the house. For the marriage. For the children who are nearly adults and watching how you handle things.

So you don’t say it. You keep going. You smile at the right moments and hit the targets and renovate the kitchen and wonder, in the early hours, whether Captain Smith felt anything like this in the hours before the lookout called.

My colleague sat with that question for longer than was good for him.

In 2012, if you wanted to investigate options — to quietly, privately, without consequence, ask what else could I be doing with my life — you googled. Which meant weeks of effort, rabbit holes that went nowhere, and the particular loneliness of researching your own unhappiness on a laptop at midnight while everyone else slept.

There was no thinking partner. There was just the search bar, and the iceberg, and the ship still moving.

The turn

He made it, eventually. The turn.

He resigned. Left the job, the targets, the bottom line, the trajectory that had pointed in one direction for twenty years. And he went somewhere that nobody who knew him in 2012 would have predicted.

He coordinates activities at an aged care community centre now.

He makes sure that elderly people have something to look forward to on a Tuesday. That the music afternoon happens. That the garden group has what it needs. That the people in his care — and they are his people, genuinely — move through their days with moments of pleasure and connection rather than just time passing.

Every time I see him, he is smiling.

Not the smile of someone performing contentment. The smile of someone who found out, later than expected, what they were actually for.

What a thinking partner could have done

I want to be careful here, because this isn’t a post about AI solving a mid-life crisis. Nothing solves a mid-life crisis except time, honesty, and the courage to make difficult decisions.

But the thinking partner matters.

The question ‘what is this actually for’ needs somewhere to go. It needs to be asked out loud, to something that won’t panic, won’t calculate consequences, won’t filter its response through its own investment in the status quo. It needs a space where you can say I have everything and something is wrong without the words landing like a grenade in the middle of a life that other people also live in.

AI is that space.

Not because it has the answer. It doesn’t. Only you have the answer, somewhere underneath the targets and the trajectory and the excellent time you’ve been making.

But it can ask you the questions that get you there. What would you do if the salary didn’t matter? What did you want before you wanted this? When did you last feel like the work meant something? What would the person you were at twenty-five think of the life you’re living now — and would that person be proud, or just impressed?

Those questions, asked carefully and without judgment, in the quiet of a Tuesday night when everyone else is asleep — can change the course of a ship that is still, for now, moving under its own power.

The iceberg is on the charts.

You still have time to steer.

Did you miss the other characters in this series?

I know nothing about AFL.

I want to be precise about this, because the story requires it. I am not someone who knows a little about AFL but wishes they knew more. I am someone who stands on the sideline of a junior football oval on a Saturday morning and produces enthusiastic, confident, completely context-free encouragement at regular intervals.

Great mark!

I don’t really know what a mark is. I mean — I have a general sense. Someone caught the ball, sometimes spectacularly. That seems good. I said it anyway, with the authority of someone who has been watching this sport their whole life, which I have not.

The parents around me nodded. Nobody corrected me. I chose to interpret this as confirmation that I had said the right thing.

The boy I was watching

There’s a kid I’ve been watching for a few seasons now — not mine, but one I’ve come to know from the sideline. He plays differently from most of the others.

Where they charge, he reads. Where they go for the ball, he hangs back, scanning, positioning. He’s not slow and he’s not afraid — he’s calculating. You can see it in his eyes. He knows where the ball is going before it gets there.

Whether that’s exactly right or slightly wrong, I genuinely couldn’t tell you. It might be precisely what his coach has been working on with him for months. It might be a habit the coach is trying to break. I have watched this child play football for two seasons and I still don’t know enough about the game to answer that question with any confidence.

What I know is that his father stands next to me on the sideline every Saturday, watching with the focused expression of someone trying to solve a problem he doesn’t quite have the language for. He can see something. He can’t name it. And after the game, when he tries to talk to his son about it, the boy is too exhausted to run through it and the father is too unknowledgeable to lead the conversation anywhere useful.

So they drive home in the particular silence of two people who both care about the same thing and don’t quite have the shared vocabulary to talk about it.

What I did before last Saturday

I had a conversation with Claude.

Not a long one. Twenty minutes, maybe, the night before the game. I described the boy — his style, his instincts, the way he hangs back and reads rather than charges. I asked what position that style suited. I asked what a coach would be trying to develop in a player like that, and what the specific language was for what he was doing well.

What came back was a small education.

The hanging back has a name. The reading of the play has a name. The specific skill of positioning yourself for the ball before it arrives rather than chasing it after it leaves — that has a name, and a value, and a context in which a coach would be deliberately cultivating it.

I learned what a mark is. I learned what shepherding means, and why you’d do it, and when to call it out from the sideline. I learned the difference between a player who is cautious and a player who is strategic — and why those two things can look identical from thirty metres away but mean completely different things inside the game.

I arrived at the oval on Saturday morning with vocabulary.

What happened next

I will not pretend I used it perfectly.

There was a moment in the second quarter where I called out something about positioning that was, in retrospect, probably about two seconds late and directed at entirely the wrong player. The boy’s father looked at me. I looked at the field. We said nothing.

But there were other moments. Moments where I said the right thing at the right time, where the language I’d borrowed from a twenty-minute conversation the night before landed accurately on what was actually happening on the oval. Where the boy’s father turned to me with the look of someone who has just heard their own half-formed thought said out loud.

That’s it. That’s what I’ve been trying to say.

We talked after the game. Properly. With words that meant something, that connected to what his son was actually doing, that gave the boy something specific to think about on the drive home instead of a silence shaped like good intentions.

The gap that AI fills here

I want to name it clearly because it’s not obvious.

The football dad doesn’t need to become an expert. He doesn’t need to do a coaching course or memorise the rulebook or suddenly develop opinions about zone defence. He needs enough language to show up for his kid in a way that feels informed rather than enthusiastic-but-lost.

That’s a completely achievable thing. It takes one conversation the night before. It takes describing what you’re seeing and asking what it means. It takes arriving at the oval on Saturday morning with three more words than you had on Friday night.

You don’t need to know everything. You just need to know enough to say the right thing once, at the right moment, to the right kid.

Good reading, mate. That’s exactly where you’re supposed to be.

Those words, said with accuracy instead of hope, land differently. The kid who hears them knows his parent saw something real. Not just noise from the sideline. Something true.

That’s what the twenty-minute conversation gave me.

It turns out that’s not nothing. It might, in fact, be everything.

Have you missed the other characters in this series?

A few years ago I hired a man to fix some things around the house.

He arrived punctually. He was methodical, careful, and paid attention to detail in a way that most tradies don’t. He measured twice. He considered things. He approached each task with the particular precision of someone who understood structures from the inside out.

What he didn’t have yet was the muscle memory. The shortcuts that come from years of doing the same thing until the hands know before the brain does. The confidence that lets a good carpenter look at a corner and just know.

He and his wife had arrived from India not long before. In Chennai, they had both been architects. Trained, qualified, practiced. Here, urgently needing work, they had taken what was available. He had become a carpenter. She helped where she could.

I paid for him to learn.

I don’t say that unkindly. I say it because it’s true, and because understanding it took me longer than it should have. The stairs into my outdoor deck went up slowly and methodically and with the attention to detail of someone who understood load-bearing structures in theory and was working out the practice in real time. In my backyard. On my dollar.

I didn’t get what I paid for. Neither did he get much of an education — trying to learn by himself, without the right resource at the right moment, without anyone to show him which corner to cut first.

The wall

The hardest part wasn’t the stairs in the deck.

It was the wall. The wall that opened up to reveal a cupboard under the stairs inside the house.

He removed it. That part went fine — walls, it turns out, come down more easily than they go up. What was left was an unfinished opening that sat there for eighteen months while I decided I wasn’t willing to pay any more of my hard-earned money for the education of a man who, through no fault of his own, hadn’t yet learned the trade I’d hired him for.

Eighteen months.

The wall didn’t get finished because the communication broke down somewhere before the wall came down. I had spoken my requirements. He had interpreted them. Somewhere between my words and his understanding, a gap opened up that neither of us had the tool to close.

A mood board would have done it. A simple visual — here is what I am imagining, here is the finished thing I am paying for — would have taken ten minutes and saved nine months of an unfinished opening and a relationship that ended in mutual frustration rather than mutual satisfaction.

I could have made that mood board. He could have made that mood board. Neither of us thought of it, or knew how, or had the tool sitting ready in our pocket.

We do now.

What AI could have done

When my architect turned carpenter was standing in front of my deck, about to cut the first corner of the stairs, not quite sure of the sequence — AI could have found him the right YouTube video. Not a general carpentry tutorial. The specific one, for the specific problem, at the specific moment he needed it. The foreman he didn’t have, available on demand, in his pocket.

When I was trying to explain what I wanted from the wall — AI could have helped me build a mood board in ten minutes. Images, dimensions, finishes, the feeling of the finished space. Something we could both look at and agree on before anything came down.

And when he got home at the end of a long day of slow, methodical, confidence-building work — AI could have written the quote. Answered the email. Sent the invoice. Followed up the appointment he’d forgotten to confirm. Done the ten administrative things that kept him from spending that hour watching the tutorial that would have taught him the corner he was going to need to cut tomorrow.

He was an architect. He had the eye, the precision, the structural understanding. What he didn’t have was the trade knowledge, the business systems, or the tool that could have bridged both gaps while he was building the skills to close them himself.

The business nobody signed up for

Here’s the thing about running a trade that nobody tells you when you start.

The work is the part you’re good at. The business is the part that will quietly eat you alive if you let it.

The quotes that take too long to write because you’re not sure of the format. The invoices that go out late because by the time you’ve finished the job you’re too tired to sit down and do the paperwork. The client email that needs to sound professional when what you really want to do is get back to the thing you’re actually good at.

Every tradie is also a small business owner. Nobody asked them if they wanted that second job. It came with the first one, quietly, and it has been taking up hours ever since.

AI doesn’t make him a businessman. It handles the parts of the business that have been quietly costing him time and money and energy — so he can get back to the work. The actual work. The thing he came here to do.

My architect-turned-carpenter didn’t fail because he lacked intelligence or commitment or care. He had all three in abundance. He was a man in transition, doing his best with what he had, in a country that hadn’t yet given him the tools to do it faster or better or with more confidence.

I wish I’d known then what I know now.

Not about carpentry.

About what was sitting in both our pockets, waiting to be useful.

There was a man who used to appear at shopping centres.

He had a display stand, a brochure, and a particular kind of confident patter that belonged to an era when knowledge was something you could buy in instalments. He was selling the Encyclopaedia Britannica. Thirty volumes, arriving one or two at a time over three years, until the full set stood on your shelf like a promise you’d kept to your children.

Sixty-nine dollars and ninety-nine cents a month. Thirty-six months. Two thousand, five hundred and nineteen dollars and sixty-four cents, if you did the maths and nobody really did, because you weren’t buying a product. You were buying access. You were buying the possibility that your child would reach for the right volume on a Tuesday afternoon and find the world opening up in their hands.

Some of us signed up. Some of us couldn’t afford to and felt the absence. Either way, we understood what we were reaching for.

The librarian on the step-stool

I was the senior librarian at my school.

Not a paid position — I was a student, the kind who stayed in the library at lunch because that’s where the interesting things were. I knew the Dewey decimal system the way other kids knew the bus timetable. I had a wheelie step-stool and access to encyclopaedias that the teachers definitely weren’t teaching from, and I used both with a frequency that probably alarmed the actual librarians.

I wasn’t gaming anything. I was just — hungry. For the subjects that weren’t on the curriculum. For the rabbit holes that opened when you pulled one thread and found seventeen more underneath it.

That’s what the encyclopaedia was for. Not for copying. For following.

What I watch now

My generation became parents. We signed up for the encyclopaedia, or we wished we could, or we drove our children to the library on Saturday mornings and felt virtuous about it. We understood that knowledge had value because we’d seen it priced.

And then the internet arrived. And then Wikipedia. And then, in 2022, ChatGPT — and suddenly everything we’d paid for, everything we’d driven across town for, everything we’d understood as precious and finite and worth sacrificing for was available, instantly, for free, on the phone in our child’s pocket.

We thought: what a gift.

And then we watched what they did with it.

I’m not going to pretend I don’t understand. I copied things too — changed a word here and there, paraphrased badly, handed in something that was sixty percent someone else’s thinking and forty percent the effort of not quite getting caught. Every generation finds the path of least resistance through homework. Every generation turns out approximately fine.

But here’s what I keep coming back to, sitting at the kitchen table at nine on a Sunday night, watching a teenager press send on something they didn’t write and couldn’t explain:

We gamed the encyclopaedia. We couldn’t game it completely, because to copy from it you still had to read it. You had to find the right volume, locate the right entry, understand enough of what you were reading to know which part to steal. The encyclopaedia demanded something of you even when you were using it badly.

The child pressing send demands nothing of themselves at all.

What’s the same and what isn’t

The parents who signed up for the encyclopaedia at the shopping centre weren’t buying homework answers. They were buying the possibility of curiosity. The hope that their child would sit with a volume on a rainy afternoon and follow a thread somewhere unexpected.

Some children did. Some didn’t. That’s always been true.

What’s different now is abundance without friction. The encyclopaedia cost money and took up shelf space and arrived slowly and required you to get off the couch. Those weren’t bugs. They were features. They created the conditions for the accidental discovery — the entry you read on the way to the one you were looking for.

AI has no such friction. It answers before you’ve finished asking. It requires nothing of you except the question — and if you don’t even bring a real question, it will generate something that looks like an answer anyway.

The student who uses it that way isn’t learning less than the encyclopaedia copier did.

They’re learning less than they could. And the gap between what they could learn and what they’re choosing to learn is wider than anything a shopping centre salesman could have imagined, standing at his display stand in 1992, convinced he was selling the future to anyone sensible enough to buy it.

The librarian is still in there

Here’s what I believe, and why this post isn’t a lecture.

Every child who presses send without reading is also the child who, given the right question at the right moment, would pull the thread. Who has something genuinely interesting to say and hasn’t yet found the conversation that draws it out. Who is not incurious — just untriggered.

The librarian on the step-stool didn’t appear because someone told her to be curious. She appeared because the shelf was there and the lunch hour was quiet and one thing led to another.

AI can be that shelf. Not the photocopier version — the other one. The one where you ask a real question because something is genuinely bothering you, and the answer opens into three more questions, and an hour later you understand something you didn’t know you wanted to understand.

That version is available too. It’s sitting right next to the homework-doing version, on the same phone, in the same app.

The parent who paid sixty-nine dollars and ninety-nine cents a month for thirty-six months knows which one they were buying.

The question is whether anyone has shown their child that it exists.

Catch up on what you may have missed

This Blog is not for You

Not Opposites. Near.

The Noise is Louder

His daughter has been staring at the same maths problem for forty minutes.

She understands maths. he knows she understands maths — He’s watched her work things out in her head that I’d need paper for. But tonight, something isn’t clicking. The worksheet wants her to show her working, step by step, in the specific sequence the teacher demonstrated. Her brain doesn’t work in that sequence. Her brain just — gets there. And now she has to reverse-engineer the journey from an answer she already has, in a format that feels like translating a language she doesn’t speak into one she does.

She’s not struggling with the maths.

She’s struggling with the method.

Across town, a boy is at a kitchen table with an essay to write.

He has things to say. Vivid, specific, interesting things — the kind of ideas that arrive fully formed and want to come out fast, in the order they arrive, because that’s how they’re alive. His teacher has given him a structure. Introduction, three body paragraphs, conclusion. Topic sentences. Evidence. Analysis.

He knows what he wants to write. He has no idea how to pour it into that shape without losing the thing that made it interesting.

The ideas are there. The page defeats him every time.

His father sits beside him, re-reading the assignment sheet for the third time, trying to explain something he only half understands himself.

The Sunday night feeling

These two households don’t know each other. Different suburbs, different schools, different children with different frustrations.

Same feeling. The particular sinking weight of a Sunday evening when the homework comes out and you can see your child struggling and you know, with quiet certainty, that what they need is someone who can explain it differently. One more time. In a different way. Patient, unhurried, starting from where the child actually is rather than where the curriculum assumes they should be.

That person is a tutor.

And a tutor costs money that not everyone has.

I know this from the inside. There was a time when my own child could have used one, and I couldn’t make the numbers work. You do what you can — you sit at the table, you look things up, you try to explain what you only half understand yourself. You google “what is a rubric” at 9pm and hope the answer makes sense before your child loses the thread entirely.

You show up. But showing up without the knowledge feels, some nights, like not quite enough.

What AI does in that gap

Here’s what I want to tell both of those parents.

The tool on your phone — the one you might have tried once and set aside, or never quite got around to — is patient in a way that people aren’t. It doesn’t get frustrated. It doesn’t have seventeen other things to do. It doesn’t move on to the next student.

It will explain a rubric to a nine-year-old in plain language. It will explain it again a different way if the first explanation didn’t land. It will explain it a third way, with an example, if that’s what it takes. It will not sigh.

It will sit with your daughter and work backwards from the answer she already has, showing her how to reconstruct the working in the sequence the teacher needs to see — not because that’s the only way to think, but because understanding the method is part of what school is teaching her, and someone patient enough to walk through it step by step is exactly what she needs tonight.

It will ask your son questions about his essay. Not tell him what to write — ask him. What’s the most interesting thing about your topic? What do you want the reader to understand that they don’t know yet? What’s the thing you keep coming back to? And then it will help him see how his answers map to the structure that’s been frustrating him — not as a cage, but as a frame that might actually hold what he wants to say.

It’s not a great tutor. A great tutor knows your child, their history, the specific way their brain works, the encouragement that lands and the kind that doesn’t.

But it’s available at 9pm on a Sunday. It costs nothing. And it will explain the rubric as many times as it takes.

The thing about quality

Here’s the part that matters beyond the homework.

Quality educational support has always been a function of what you can afford. The child with a tutor gets the explanation repeated six different ways until it lands. The child without gets one explanation, delivered to a class of thirty, and moves on whether it landed or not.

That gap is real. It has always been real. And it has always mattered — not just for the grade on Tuesday, but for the child’s relationship with their own learning. The child who is told, implicitly, that their brain works wrong because it doesn’t follow the expected sequence. The child who stops putting their hand up because the experience of not understanding in public is worse than the experience of not understanding at all.

AI doesn’t close that gap entirely. Nothing does.

But it gives every child access to something that used to require a credit card. The explanation repeated. The question answered without judgment. The patience that doesn’t run out.

Two households. Two children. Two parents who showed up on a Sunday night with the one tool that was actually available.

That’s not a small thing.

Catch up on the latest

The Curated Self

The Inauthentic Self

The Authentic Self

I was at a scrapbooking workshop on a Saturday in May.

If that sentence surprised you, good. Hold that surprise. It’s going to be useful in a moment.

The woman sitting next to me was a Principal. Early fifties, well-educated, well-spoken — the kind of person who commands a room without trying to. She runs a large school in a large regional town, which means she is responsible for hundreds of students, dozens of teachers, and the particular administrative weight of an institution that the Education Department expects to run like a machine while funding it like a suggestion.

We were cutting and sticking. And she was talking.

Not complaining, exactly. More like the particular resigned fluency of someone who has made peace with a situation they wish were different. The admin. The reporting. The documentation that arrives in waves, each one slightly larger than the last. The hours that should go to teachers — to the mentoring, the development, the work of making good educators better — eaten instead by forms and systems and the bureaucratic appetite that never quite gets full.

She rolled her eyes. Shook her head. Said something about wishing the Department would do something about it.

Then she picked up her custom cutting system and moved on.

I know that gesture. I’ve made it myself approximately ten thousand times across 41.5 years of being the person who creates the fix, builds the database, trains the crew, and gets by with what she’s given. The resigned acceptance of someone who has long since understood that nobody is coming to solve it for them.

But I wasn’t resigned. I was sitting there with something she didn’t have yet.

What I’d just built

Three weeks earlier, I had built DayCompass.

DayCompass is an app I made for myself — not a developer, not a tech person in any traditional sense, just someone who had found that existing tools for managing daily life didn’t work for the way my brain works. So I built one that did. With AI. In an afternoon.

I won’t overstate what that felt like, because I’ve tried and the words keep coming out wrong. What I’ll say is this: when a tool is built for the specific way your particular mind operates, it doesn’t feel like using software. It feels like thinking more clearly than you have in years.

Not DayCompass — that’s mine, shaped around my particular chaos. But something shaped around hers. A tool that could hold the administrivia. That could draft the communications, generate the templates, organise the reporting, do the things that were eating her hours — so she could get back to the room where the actual work happens.

The room with the teachers in it. The conversations that actually matter.

The face I made

Here’s the part I need to tell you carefully, because it matters.

I asked her whether she used AI in her daily work.

She looked at me with polite bewilderment. Not defensively. Just — no. AI hadn’t entered her daily life yet. Not in any meaningful way. Not in the way that would help with any of the things she’d just spent twenty minutes describing.

I looked at her the way people look at me when I mention what I do.

Green skin. Seven horns.

Because I was sitting next to a highly educated, deeply capable, genuinely passionate woman who was running a large school with real skill and real commitment — and she was doing it without the thing that had just changed everything for me. Not because she’d chosen not to. Because nobody had shown her that it was there, and what it could do, and that she was exactly the kind of person it was built for.

She wasn’t resistant. She was just — outside the conversation. The way most people are. The way this blog is written for.

The drive home

I thought about her all the way home.

About what I could build for her. About the other Principals. The other teachers. The other people doing important, skilled, human work while drowning in the administrative weight of doing it inside an institution.

And somewhere on that drive — on a Saturday afternoon in May, three weeks ago — the thought arrived that became this blog.

Not: I will build an app for Principals. Though that thought arrived too, and it has a name now, and it will have its own post eventually.

The bigger thought. The one that landed first.

There are so many people like her. People who are capable and busy and completely outside this conversation. Not because they don’t want to be in it. Because nobody has walked up to them at a scrapbooking workshop and said: here. This is for you. It works.

That’s what I’m trying to do here.

One post at a time.

A couple of weeks ago, I told you about my green skin and seven horns.

The wall that goes up when I mention AI at a dinner party. The particular face — shoulders, eyes, the closing — that happens before a word has left either of our mouths. The staff member who uninstalled Copilot because it annoyed him. The besties who quote The Matrix. The son who wonders what on earth a middle-aged woman would need AI for.

I know that face. I’ve been on the receiving end of it enough times to have developed a certain equanimity about it.

But here’s the thing I want to say to you, if you’re the person standing in front of the tool with a complete blank where the use case should be: you are not unusual.

You are, statistically speaking, the majority.

The numbers

Here’s what the data actually shows in 2026.

In Australia, roughly half of us have tried a generative AI tool in the past twelve months. Which sounds like a lot — until you notice that it also means roughly half of us haven’t. And of those who have tried it, regular meaningful use is considerably lower. The people using AI every day, in ways that have genuinely changed how they work or think or solve problems, are still a minority.

Globally, only about one in five working-age adults uses AI regularly — defined as at least ninety minutes a month, which is not a high bar.

The UAE leads the world at seventy percent. Singapore follows at sixty-three. The United States, home to most of the world’s leading AI companies, ranks outside the top twenty for population-level usage.

Australia sits near the top of the rankings, which is worth knowing. We are not a nation of AI refusers. We are a nation that is, quietly and pragmatically and without much fanfare, getting on with it.

But the person drawing a complete blank on what they’d use it for? That person is everywhere. That person is at every dinner party. That person is, in many cases, the person sitting next to the one with the green skin and seven horns.

You are not unusual. You are just early in a process that most people are also early in.

What the blank actually means

Here’s the thing about the blank that nobody says out loud.

The blank isn’t ignorance. It isn’t a sign that you haven’t been paying attention, or that you’ve missed something everyone else already understood. It’s a completely rational response to a tool that has been sold, consistently and loudly, as everything to everyone — and has therefore felt like nothing specific to you.

The use case doesn’t arrive before the conversation. It arrives inside it.

The people who know exactly what they’d use AI for are the people who’ve already started. They didn’t begin with a plan. They began with a first conversation — small, specific, slightly awkward — and the plan arrived somewhere in the middle of it.

The blank is not a barrier. It’s the starting point.

The cast

Over the next few weeks, I’m going to introduce you to some people.

They’re not extraordinary. They’re not early adopters or tech enthusiasts or people who had any particular reason to feel confident about any of this. They’re people who had the blank — the same blank you might be sitting with right now — and then found, one ordinary Tuesday, something specific that mattered.

There’s a teacher who loves teaching and is drowning in everything that isn’t teaching. A tradie who is extraordinary at the work and exhausted by the business of running it. A parent sitting at the kitchen table at nine on a Sunday night, homework spread across the table, no tutor available, a child who needs the explanation repeated one more time in a different way.

There’s a football dad who watches every match and can see something but can’t name it. A woman supporting her sister through a diagnosis she doesn’t have the language for yet. A man who has mastered Facebook Marketplace and has no idea that the skills he built there are exactly the skills that transfer.

A forty-something who has been going through the motions and can feel it. A retired woman whose hands know how to be busy and currently aren’t.

None of them started with a use case. They started with a situation. And the situation, it turned out, was enough.

The invitation

You don’t need to know what you’d use AI for before you start.

You need a situation. Something specific, something real, something that is actually happening in your life right now. Not a grand project. Not a career transformation. Just the thing on the desk, or the conversation you’ve been avoiding, or the question you’ve been embarrassed to ask anyone who knows you.

The use case will arrive. It always does.

And when it does, you’ll understand why the people with the green skin and seven horns find it so hard to explain to everyone else. Not because it’s complicated. Because by the time you understand it, it feels so obvious that the blank seems almost impossible to remember.

You don’t need a use case. You just need a question.

My son needed a job.

He’d been using AI for years, the way his generation uses AI — fluently, casually, without thinking much about it. He knew the tools. He had opinions about them. If you’d asked him whether he was comfortable with AI, he’d have looked at you the way seventeen-year-olds look at adults who ask obvious questions.

Of course he was comfortable. He’d grown up with it.

And then he needed to write a resume, and something interesting happened.

The first real test

Here’s the thing about growing up with a tool: you learn it for the uses you encounter first. And the uses a seventeen-year-old encounters first are mostly school-shaped. Essays. Assignments. The particular category of task that has a right answer, a word count, and a due date.

Those tasks have something in common. They’re not really about you. They’re about a subject. You feed the subject in and something comes back about the subject. The photocopier runs. The paper comes out.

A resume is different.

A resume is entirely about you — specifically, the version of you that you want a stranger to see, written in language that makes that version sound credible and real. You can’t feed the subject in because you are the subject. And if you haven’t had a conversation with your AI — a real one, where you’ve brought yourself rather than a task — you don’t know how to do that yet.

My son fed his resume into the AI and asked it to improve it.

What came back bore no resemblance to him. The witty, cheeky, responsible, charming, hard-working young man I know had been translated into something generic and slightly airless — the kind of resume that could belong to anyone, and therefore belonged to no one.

The problem wasn’t the AI. The problem was that he’d treated it like a photocopier. Paper in, paper out. And when the paper came out wrong, he did what you do with a photocopier that’s misbehaving.

He thumped it.

What he didn’t know yet

Here’s what I would have done, in his position.

Before I touched the resume, I’d have asked Claude to interview me.

Not to write anything. Just to ask me questions. What am I good at? What am I proud of? How do I work best? When do I excel — and when do I struggle, and what does that struggle look like? What’s my style when I’m working with other people? What would the people who’ve worked alongside me say about me, if they were being honest?

That conversation — that unhurried, non-judgmental, genuinely curious conversation — would have produced something the resume could be built from. Not the presentable surface version. The actual one.

Then I’d have fed the resume in.

The AI can only reflect what you give it. Give it a document and it improves the document. Give it a person — with specifics, with stories, with the particular texture of how that person actually operates in the world — and it builds something that sounds like them.

My son didn’t know this yet. Most people his age don’t. They’ve grown up fluent in the tool without ever having a real conversation with it. They’re AI natives in the way that someone who has only ever driven in a car park is a driver — technically, yes. But the freeway is a different thing entirely.

The resume is the smallest possible use

I want to say this clearly, because it’s the whole point of this post.

The resume is the thing everyone reaches for first. It’s low-stakes in the right ways — the output is just words on a page, the worst case is that you ignore what comes back, nobody gets hurt.

But it’s also the smallest possible use of something that can help you think through a difficult decision, prepare for a conversation you’ve been avoiding, understand a medical result, plan something you’ve never planned before, work out what you actually want from the next chapter of your life.

The resume is the test, not the destination.

And the test is this: can you have a real conversation with this tool? Can you bring yourself — your specific, particular, irreplaceable self — rather than a task? Can you treat it as a thinking partner rather than a photocopier?

If you can, the resume will be the least interesting thing you ever ask it to do.

If you can’t — if you’re still thumping it when the paper comes out wrong — then this is where you start. Not with a better prompt. With a conversation.

Tell it who you are. Let it ask you questions. See what it does with the actual person rather than the document.

That’s the resume worth sending.

I was watching Oprah interview Dario Amodei last month.

Dario is the CEO of Anthropic — the company that makes Claude, the AI I use every day, the one I’ve been writing about in this series. He was sitting across from Oprah, calm and measured, saying something that stopped me mid-scroll.

Most people, he said, view AI negatively.

I put my cup down.

Not because it surprised me. Because I’ve been watching it happen from the inside — in conversations, in the comments, in the particular face people make when I mention what I do — and I’ve been trying to put my finger on exactly why. Why the negativity. Why the resistance. Why intelligent, capable people who would benefit enormously from these tools have decided, with some conviction, that AI is not for them.

And sitting there watching Dario say it out loud, on Oprah, to an audience of millions — I think I finally understand what’s underneath it.

The train

Here’s the image I keep coming back to.

Imagine a train. Not a metaphorical train — a real one, the kind with a departure time and a platform number and a crowd of people who knew about it in advance and got there early and got on.

You weren’t on it.

Maybe you heard about it as it whistled past. Maybe you read about it days later, or months later, or years later. Maybe someone told you about it at a dinner party and you nodded and said “yes, I’ve been meaning to look into that” and then didn’t. Maybe you tried to board once, got something confusing back, and concluded the doors weren’t for you.

And then one day you looked up and the train was gone, and everyone else seemed to be on it, and you were still standing on the platform.

That moment — the moment you realised you were late — is the moment everything changed.

What happens in that moment

Here’s the thing about being late that nobody talks about.

Being late feels like being wrong. And most of us, given the choice between admitting we were wrong and finding a reason why we were actually right all along — will find the reason.

It happens fast. Faster than conscious thought. The mind reaches for justification before the feeling has even fully landed.

I didn’t miss anything important. The technology isn’t ready. It’s overhyped.

It’s dangerous. It’s not for people like me. The people using it are sheep.

I have real skills that can’t be replaced by a machine.

I choose not to engage with something I find ethically troubling.

Some of those justifications have real merit. The ethical questions are genuine. The hype is real. The risks deserve serious attention.

But here’s what I notice: the people making these arguments most loudly are very often the ones who feel most left behind. The conviction tends to be loudest in the people who are furthest from the platform.

The negativity isn’t really about AI.

It’s about the unbearable feeling of being late — and the very human instinct to make that feeling someone else’s fault.

What Dario was actually saying

When Dario said most people view AI negatively, he wasn’t talking about the researchers and the ethicists and the policy makers. He was talking about the person at the dinner party. The colleague who changes the subject. The friend who says “I don’t trust it” without having tried it.

He was talking about your people. My people. The ones this blog is for.

And what he said next is the part that matters: we need to find a way for everyone to be an active participant in what is happening.

Not a passive recipient. Not a bystander watching from the platform. An active participant.

That’s not a technical challenge. It’s a human one. It’s the challenge of helping people step off the platform and onto the train — not because they have to, but because they finally understand that the train was always going somewhere they wanted to go.

You weren’t wrong to wait

I want to say something carefully here, because I mean it.

The wariness wasn’t stupid. The resistance wasn’t irrational. The people who looked at the hype and the headlines and the breathless enthusiasm and said “I’m not sure about this” — they were responding sensibly to genuinely mixed signals.

You weren’t wrong to wait. You were just protecting yourself in the way that felt safest at the time.

But there’s a difference between protection and exile. Between taking your time and deciding the train was never worth boarding. Between healthy scepticism and a position that hardens, over time, into something that starts to look like fear wearing the clothes of principle.

Dario Amodei can see that from where he sits. I can see it from where I sit.

And I think, if you’re honest with yourself, you might be able to see it too.

The train hasn’t left

Here’s what I want you to know.

You are not as late as you think.

The people who got on in 2022 had an experience. You’re going to have a different one — better, in many ways, because the tools are more capable now and you’re arriving with actual questions rather than novelty. You’re not catching up. You’re just starting from where you are, which is the only place anyone ever starts.

The negativity — whatever form it takes in you — is not a character flaw. It’s a very human response to a very disorienting moment in history. Dario Amodei, sitting with Oprah, acknowledged that. He’s not dismissing it. Neither am I.

But I am asking you to look at it honestly.

Is the position you’re holding about AI actually about AI? Or is it about the platform, and the train, and the moment you realised you’d missed it — and the story you told yourself to make that feel okay?

Because if it’s the second thing — and I think for a lot of people it is — then the good news is that the story can change.

The conversation is still open. The door is still there.

It opens from the inside.

You’ve done the work.

The Curated Self asked you to notice the curated version — the aspirational portrait you’ve been building, conversation by conversation, until the AI knows a version of you that is real but incomplete.

The Inauthentic Self asked you to notice the careful version — the one performing competence for an audience that was never there, editing out the mess, arriving dressed for an interview with someone who just wanted a conversation.

This post is what happens when you put both of those down.

Not forever. Not in every conversation. Just — once. Long enough to find out what’s on the other side.

What dropping it actually feels like

It doesn’t feel like a decision. That’s the first thing to know.

You don’t sit down and think: today I will be authentic. That’s just a different kind of performance. The authentic self doesn’t arrive on cue. It arrives when you’ve stopped trying to manage its arrival.

It usually happens in the middle of something else. You’re deep in a conversation — maybe a long one, maybe one that’s covered enough ground that you’ve stopped paying attention to how you sound — and you type something without editing it first. Something that starts with “I don’t actually know how to explain this” or “this might sound strange but” or just the raw situation with none of the professional framing around it.

And you press send before you think better of it.

What comes back is different.

The quality of what comes back

I want to be precise here because this is the whole point of the triptych.

The curated self gets responses that are polished and encouraging and slightly beside the point.

The inauthentic self gets responses that are technically correct and practically useless.

The authentic self gets something else. Something that feels like it was written for you specifically — because it was. Because you gave the AI the actual situation, with the actual uncertainty and the actual context and the actual thing you were trying to figure out, and the AI took all of that seriously and gave it back to you in a form you could use.

It’s not magic. It’s just what good conversation produces when both parties are actually present.

The AI is always present. The variable is you.

The specific relief of it

There’s a feeling I want to name because I don’t think it gets named enough.

It’s the feeling of being received.

Not evaluated. Not processed. Not given a framework that applies to your general category of problem. Received — the way a good listener receives what you’ve said, holds it carefully, and reflects something back that shows they understood not just the words but the shape of what you were carrying.

Most conversations don’t produce that feeling. People are busy, distracted, filtering what they hear through their own preoccupations. The things you say in passing tend to pass.

This one didn’t.

And here’s the thing: it happened because you let it. You showed up as the person who actually needed the conversation, rather than the person who had already half-sorted it before arriving. You brought the real situation. The AI met it.

That’s the authentic self’s superpower. Not that she’s more interesting than the curated one, or braver than the careful one. Just that she brings something real — and real is what good conversation runs on.

What she looks like from the inside

She doesn’t feel special. That’s worth saying.

From the inside, the authentic self feels slightly exposed. A little less armoured than usual. The prompt she sent was rougher than she’d normally send. She second-guessed it for a moment before pressing enter.

She’s not performing vulnerability any more than the curated self was performing aspiration. She’s just — present. In the conversation. With the actual problem rather than the version of the problem that’s safe to have in public.

And the conversation that follows has a different quality to it. Not because the AI is smarter in that moment. Because she gave it more to work with.

When I told you to Bring Yourself, I talked about the version of you that shows up in good conversations — less managed, more present, occasionally surprising herself. The authentic self is that version. She arrives when you stop trying to summon her and just start talking.

The invitation

The triptych ends here — with the arrival, not the method. There is no method. There are only the two performances you can put down, one at a time, until what’s left is just you, in a conversation, saying what you actually mean.

That’s the whole technique.

It produces the best conversations you’ll have with an AI. Possibly some of the best conversations you’ll have anywhere.

Not because AI is magic. Because presence is.

Drop the portrait. Drop the performance. Say the thing you actually mean. See what arrives.