The Dyslexic AI Newsletter by LM Lab AI

Some mornings you wake up and the first clear thought in your head is: what the fuck am I doing with my life.

Not a crisis. Not a breakdown. Just that flat, honest question sitting there before your feet hit the floor. Why do I keep doing this to myself. Is any of it worth it.

That was this morning. That’s still where I am right now, dictating this.

I’ve put the last two years into this. The previous business, the savings, the time, all of it went into Dyslexic AI and LM Lab and the Cognitive Partner work. Some days I look at that and see a body of work.

Today I look at it and see two years I might not get back.

So no lesson today. No research paper, no framework. Just the actual view from inside this thing, because I’ve decided that pretending is worse.

Here’s what I scrolled past before I even got out of bed.

Half my feed says AI is a catastrophe. The data centers are the enemy. The whole thing is theft. And every one of those posts gets typed on a phone, uploaded through those same data centers, in between songs served by a recommendation engine, next to a voice assistant nobody thinks twice about.

Some of those concerns are real and worth arguing about. I’m not dismissing them. But the certainty doesn’t survive five minutes of looking at how the person posting actually lives.

The other half is the exact opposite. Someone cracked it. Revenue screenshots. Launch numbers. Six figures off a workflow they threw together on a Saturday. Every single one of them looks like they found the door while I’m still running my hands along the wall.

Both sides are selling certainty. The real work happens in the boring, unglamorous middle, and nobody posts from there because there’s nothing to sell.

I’ve been living in that middle for two years. It is not photogenic.

Let me be precise, because precision is about all I’ve got today.

AI did not make everything easier. Sometimes it makes things harder. Sometimes I hand it a simple problem and it hands me back a gorgeous, elaborate, over-engineered version of that same problem, and I lose an afternoon to it.

But it also showed me my own limits with a clarity I’d never had before. And then it let me work straight past several of them. I have built things in the last two years that I flatly did not believe I was capable of building. That’s not a marketing line. That is just what happened.

So it gave me abundance. Real abundance. Every day I see three more things that are now possible for me that weren’t before.

An abundance of possibility is not the same thing as an abundance of income.

That’s the whole trap. AI will hand you a hundred open doors and not one of them comes with rent attached. Both of those things are true at once, and the second one does not make the first one fake.

Nobody puts that in a carousel post. It doesn’t convert.

Here’s the part that actually belongs in this newsletter.

If you’re dyslexic, you got trained early. Trained to measure yourself on a scoreboard that was never measuring your real work. Spelling test. Reading level. Timed exam. You could be doing the hardest, most original thinking in that room and still sit dead last on the chart.

You learn something from that, and it sticks. You learn the chart is the truth and you are the problem.

Then you grow up. You get out. You think you’re finally done with it.

And the feed hands you a brand new scoreboard. Followers. Revenue. Launch day. Who’s winning at AI this month.

If you spent your childhood being measured wrong, you are unusually easy to measure wrong again as an adult. Same mechanism. Same sick feeling in your stomach. Different chart.

I’m not going to tell you to ignore it. I can’t do that myself and I’m not going to hand you advice I don’t follow. But every time that number comes at you, ask what it’s actually counting. And whether it has the first clue what you’ve been doing.

Mine doesn’t. Yours doesn’t either.

Here’s the thing I worked out this morning, and I only worked it out because I said it out loud to a machine at six in the morning like a lunatic.

I started this newsletter to get all of this out of my head and put it somewhere people like me might find it. Three hundred and seventy editions later, I’ve done exactly that.

And it has gone almost entirely one direction.

I tell my kids constantly: go talk to a person. Don’t just talk to the tools. And then I turn around and build my entire working life around talking to the tools, because I still haven’t found my room. My people. My team. So I have the conversation with myself and I publish it and I hit send.

I have gotten real notes from real readers. Thank-yous that have gotten me through weeks I’d rather not describe. But a thank-you every so often is not the same as a room full of people building next to you.

And here’s the part that’s my fault, not yours.

In 370 editions I have never once asked you for anything. No pitch, no ask, no request. Everything free, always. And then I got quietly bitter that it stayed one-way.

That’s on me. I’m fixing it today.

One thing this week. Ten seconds.

Hit reply and send me one sentence. Either something you’re working on that hasn’t paid off yet, or one thing from this newsletter that ever actually helped you.

If replying to a newsletter feels weird, email me straight at [email protected] and put STILL HERE in the subject line. That one goes to me, not a queue, and that subject line means I’ll find it.

One sentence. That’s it. I read every one myself.

And if you’re having your own version of this morning: you’re probably not behind. You’re just not posting about it.

For the next day it feels like none of it added up.

“I’m having a day where I feel like I have nothing to show for the work I’ve been doing. Before you say anything encouraging, help me build an honest ledger. Ask me questions to pull out what I’ve actually built, learned, changed, or gotten better at over the last two years, including the things I don’t count because they didn’t make money. Then tell me plainly what the ledger shows, including the real gaps. Don’t cheer me up. Just be accurate.”

The point is not to feel better. The point is to see the real list instead of the one your worst hour wrote for you.

I’m still here. Still building. Still doing this in public, which is a strange way to live, but it’s the only honest way I know how.

I'm putting together a real pulse check on where dyslexic thinkers actually are with AI in 2026. What you're using, how often, what still isn't working. It goes out in the next edition, and your answers decide what I build next.