Edition 366½ | The Dyslexic AI Newsletter by LM Lab AI A special edition — unscheduled, undictated-from-a-desk, happening right now.

What You'll Learn Today

Reading Time: 6 minutes | Listening Time: 8 minutes

I am dictating this right now.

Not typing. Not editing. Talking.

And somewhere in the background, the same AI that just fixed my newsletter is turning what I say into the edition you are reading right now.

That sentence used to be science fiction. This morning, it was just Tuesday.

You might be wondering about the 366½.

I have the next several editions already written and ready to go. I am not going to renumber them.

But what happened this morning was too good to hold until the next scheduled slot. So this is a half edition. A bonus drop between the regular ones. Something that happened in real time and needed to be shared in real time.

That is the whole point of it, actually.

Let me back up.

For the past two years I have been trying to build a workflow that actually works from end to end. Not just a piece of it. Not "I can dictate a rough draft and then go fix the formatting myself."

All of it. Start to finish. Idea to inbox.

I have had various versions working at various times. Some of them were impressive. Most of them had a gap somewhere — a moment where I had to sit down, pick up a keyboard, and do the thing that my brain makes harder than it should be.

Over the last couple of months I stepped back from experimenting. Focused on content. Summer. The kids. The stuff that matters more than the tools.

But I kept thinking about that gap.

This morning, I closed it.

Here is what happened.

I had Newsletter 366 ready to send — the one about the four clusters of human ability, the one I finished dictating from a bench at Disneyland while my kids were on a ride.

The writing part worked great. The formatting did not.

When the draft came into beehiiv, it arrived as one giant blob of text. Paragraph breaks missing. Section headers buried inside body paragraphs. The kind of visual chaos that makes a newsletter impossible to read on a phone — which is exactly how most of my readers consume it.

Fixing that manually would have taken me 45 minutes. Maybe more.

That kind of detail work — hunting for missing line breaks, reformatting headers, cleaning up spacing — is the part of publishing that my brain finds genuinely exhausting. It is not the hard thinking. It is the mechanical translation layer. And it costs more than people realize.

So I described the problem out loud. To Claude.

Not "here is a document, please fix it." I talked through what I needed — find Newsletter 366, confirm the Disneyland version is in there, check the formatting, fix it.

Claude connected directly to my beehiiv account, pulled up the draft, read it, and told me exactly what was wrong.

Then I watched.

Claude opened my beehiiv editor in Chrome. I could see the page load, the content change, the sections being restructured in real time. Every section header became a proper heading. Every paragraph got its own breathing room. The TL;DR at the bottom went from a wall of text to a clean bullet list.

I did not touch the keyboard once.

When it was done, the newsletter looked exactly the way I would have wanted it to look if I had spent an hour doing it myself.

Except I had been sitting on my phone. Talking.

I want to be honest about why this matters beyond the cool factor.

For most people, reformatting a newsletter is annoying.

For a lot of dyslexic creators, it is a genuine barrier. Not a small inconvenience. A real obstacle that sits between the idea and the audience.

The ideas have never been the hard part. 366 editions in, the ideas keep coming. That has never been what slows me down.

What slows me down is the translation layer. Getting the idea out of my head, through the right tools, into readable shape, formatted correctly, and into people's inboxes — without losing so much energy on the mechanics that I have nothing left for the thinking.

What I experienced this morning was that translation layer shrinking to almost nothing.

And here is the part I want you to sit with for a second.

I am not using a keyboard right now. I never opened a text editor. I described what I needed the way I would explain it to a smart colleague sitting next to me, and the work got done.

That is not just convenient. That is a fundamentally different relationship with the tools.

For years, the workflow assumed you could type, edit, format, and sequence information quickly and accurately. Those assumptions were baked into every platform, every publishing tool, every newsletter editor.

They were not built for how our brains work.

What is changing now is not just that AI can do more tasks. It is that the interface between your brain and the tools is finally starting to meet you where you are — instead of making you meet it where it is.

Voice. Conversation. Describing what you need instead of executing it step by step.

That shift matters enormously for dyslexic and neurodivergent creators. Because the gap it closes is not a small efficiency gain. It is the gap that has been costing us energy we never had to spare in the first place.

OK But What Do I Actually Do With This?

One thing to try this week.

Pick the single mechanical task in your content process that drains you the most. The one you dread. The one that sits between the idea and the finished thing.

Then describe it out loud to Claude — not as a command, but as an explanation. Talk through what you need the way you would explain it to a smart person who just walked into the room.

See what happens.

The goal is not to automate your creativity. It is to stop spending your creative energy on things that are not creative.

Steal This Prompt

Use this when you have content that came out formatted wrong.

"I have a newsletter that I need cleaned up for mobile reading. My readers are often dyslexic or neurodivergent, so I need short paragraphs with lots of space between them, clear section headers, and bullet lists instead of long walls of text. Here is the content: [paste or describe it]. Go through it and reformat it so it breathes on a phone screen."

Then watch it happen.

Matt "Coach" Ivey Founder, LM Lab AI | Creator, The Dyslexic AI Newsletter Dictated, not typed. Obviously.

Previously

Next Edition 367: "The Wave Model of Meaning" — Modern AI does not work like a light switch, on or off. It works more like a field of relationships, meaning emerging from context and pattern. And that connects directly to how a lot of dyslexic minds already describe their own thinking.