How I Write My Articles by Talking to My Phone

Most of my articles don't start in front of a blank page.

They start anywhere.

In the street. On my couch. Walking. At a café table. Right after a meeting. Sometimes at the very moment I thought I was doing something else.

An idea comes back. It insists. It isn't an article yet. Not even an outline, necessarily. It's a tangle of mixed things: a hunch, an irritation, an example, a memory from work, a sentence that keeps returning, a contradiction I still can't quite put into words.

I used to tell myself: I'll write it later.

Then later arrived, and the idea had lost its force. The subject was still there, but not the momentum. The precise example had vanished. The sentence that held it all together had evaporated.

Today I do it differently.

I talk to my phone.

I Empty My Head

The first move isn't to write.

The first move is to empty my head.

When an article idea shows up, I don't immediately look for:

• the right title;
• the right outline;
• the right introduction;
• the right phrasing;
• the right conclusion.

I talk.

I say what I've understood, what I've lived through, what annoys me, what strikes me as interesting. I can repeat myself. I can start from the end. I can mix two ideas together. I can say a clumsy sentence and correct it thirty seconds later.

That's fine.

At that moment, I'm not drafting.

I'm getting the material out.

That's the whole point of the setup: not asking thought to be clean too early. An article demands structure, progression, poise. But the idea itself often arrives in a much rawer form.

The phone lets me respect that moment.

I can talk without ordering.

The Phone Is Wired Into the Real Workspace

The phone here isn't just a voice recorder.

That matters.

When I talk, my words don't end up in some isolated note I'll have to dig out later. My phone is wired into my computer. My computer is wired into an AI that can work inside my files. And that speech is added directly to the session source of the right context.

The Mac works. Codex executes. The AI reads and writes in the right folder.

Me, I can be somewhere else.

That's what makes this workflow different from ordinary dictation. Dictation captures speech. My system brings that speech into a workspace that already has a goal, sources, notes, drafts, sometimes decisions or earlier texts.

Capture isn't only a backup.

It's already a positioning.

It's Not Just Swapping the Keyboard for Voice

You might think this workflow amounts to using a mobile word processor.

That would miss the point.

A word processor receives what I write. It can be very good. It can sync my files, transcribe my voice, let me correct things and move paragraphs around.

But it stays passive.

It doesn't ask me what's missing.

It doesn't see that an idea is fragile.

It doesn't notice that I forgot the central example.

It doesn't remind me that I've already written something close to this.

It doesn't turn a spoken hunch into a working question.

With the phone wired into the context, speech isn't merely captured. It immediately becomes material the AI can interrogate.

The text isn't just waiting to be reread.

It's starting to be worked on.

Nor Is It AI at the Desk

I could do much of this work at my computer.

Open the AI. Explain the idea. Ask for objections. Build an outline. Draft.

I do that too.

But it isn't the same moment.

Sitting at the desk assumes I'm already settled into a work session. You have to be at the screen, available, ready to organize your thinking.

And a lot of ideas aren't born in that state.

They arrive when the body is doing something else:

• walking;
• on public transport;
• coming out of a call;
• at a café table;
• in the evening, when the subject keeps turning;
• at the moment I hadn't planned to work.

If I wait until I'm back at the computer, I usually keep the subject but lose the temperature of the idea.

Temperature is the energy of the moment: the exact example, the connection that just appeared, the reason the subject matters to me right now.

The phone lets me catch the idea while it's hot.

The context keeps it from staying a pile of lost scraps.

The AI lets me start questioning it right away.

Most Articles Ship Fast

I also work this way because I want most of my articles to ship fast.

Not in six weeks.

Not after building some heavy apparatus around the subject.

In my case, they need to be able to ship within two days at most.

That doesn't mean everything is always published in two days, or that the first draft is already the final article.

It means the essentials have to get out of my head very quickly.

I can dump:

• what I already know;
• what I've just understood;
• the examples that come back to me;
• the objections I can sense;
• the limits I want to keep;
• the sentences that carry the hunch.

That short window matters.

It gives the workflow its rhythm. I don't need to build a large documentary base up front. I usually start from a recent experience, a problem I ran into, a conversation, a change of mind.

The material is still in my head.

It has to come out before it settles.

The Session Source Keeps the Raw Material

When I talk like this, I'm not producing a clean note.

I'm producing a session source directly.

It's raw material: spoken, disordered, sometimes repetitive, sometimes incomplete. It keeps what I say, but also what I clarify, correct, confirm or reject in the exchange with the AI.

This session source is allowed to be messy.

It should even be a little messy.

If it gets too clean too early, it loses part of its value. It starts to look like a draft, when it should be the primary material the draft is made from.

A published article has to be clear.

The session source has to be faithful to the moment the thinking came out.

It keeps the raw material so the work can begin.

AI Surfaces What's Missing

Once the material is down, I don't immediately ask the AI to write the article.

That would be too fast.

I first ask it what's missing.

That's often the most useful moment.

When I write alone, a lot of the gaps show up later. I reread the text two days on. I go for a walk. I take a shower. An objection surfaces. I suddenly see that an example is missing, that an idea moves too fast, that a distinction isn't clear.

That moment is precious.

But it often comes late.

With AI, it can come in ten seconds.

It can surface:

• a blind spot;
• a contradiction;
• an absent example;
• a fuzzy distinction;
• an unexplained assumption;
• an idea too weak to stand on its own;
• an appealing thread that's off topic.

And inside a context, that feedback is more interesting than a generic list of objections.

The AI can compare my material with what already exists:

• earlier texts;
• notes;
• session sources;
• decisions already made;
• angles already ruled out;
• the goal of the article.

It can see that a passage repeats another article, that an example doesn't really support the thesis, that a conclusion goes further than what I actually described.

It doesn't replace thinking.

It speeds up the appearance of what's missing.

Only Then Do I Build the Text

Next comes the outline.

The outline isn't the starting point.

It arrives after the raw material and the first objections. That's why it's better. It isn't trying to organize emptiness. It's organizing something that has already started to live.

I ask for a structure. Then I correct it. I move a section. I cut one. I add another. I say the article should open with a scene, or with a thesis, or with a contradiction.

Two or three iterations are usually enough.

At that stage, the article almost exists.

Not yet in its sentences.

But in its logic.

When I finally ask for a draft, the AI isn't starting from nothing. The ideas are there. The examples are there. The objections are there. The trade-offs are there. The outline is there.

The AI turns recent, disordered, spoken thought into readable text.

That's an enormous amount of work. When you talk at length, you repeat yourself, you double back, you mix two ideas, you sometimes give an example before you've stated the problem.

Reworking all that by hand is a slog.

The AI groups, cuts, rephrases, smooths, finds the thread again.

I don't see why I'd deny myself that.

Why It's Still My Article

The question comes up often: if AI does the writing, is it still my text?

For me, the answer depends on what you delegate.

If I delegate the idea, the angle, the experience, the examples, the judgment and the responsibility, then no, it isn't really my article anymore.

But if the AI turns my material, my reasoning, my choices and my corrections into readable text, then the article stays mine.

What matters isn't that every sentence comes straight from my fingers on a keyboard.

What matters is that the ideas, the organization, the nuances, the angle and the choices come from me.

After the drafting, I reread. I put a nuance back in. I cut a sentence that sounds too certain. I add an example. I reject a sequence that rings false.

The AI speeds up the transformation. It doesn't replace the judgment calls.

What This Workflow Changes

This workflow has changed my relationship to writing.

Before, writing usually meant finding an available slot, a desk, a keyboard, enough concentration, and then starting again from scratch or nearly.

Now the article can start when the thought arrives.

The real change comes from the combination:

• the phone captures at the right moment;
• the computer gives access to the real workspace;
• the context files the speech into a session source;
• the AI surfaces what's missing;
• the dialogue clarifies the idea;
• the outline sets the progression;
• the drafting transforms;
• the reread makes the text truly mine.

Taken separately, each of these pieces already exists.

Together, they change the nature of the work.

It isn't only writing faster.

It isn't only dictating.

It isn't only chatting with an AI.

It's bringing a still-warm thought into a system able to keep it, question it and transform it.

And for an article, that's often exactly what I need.