Using AI with ADHD changed how I start things. I can dump a messy, too-big idea on it, and it figures out where I’m going and hands back a small first version I can begin today. Before, I’d make every idea so big in my head that I’d get overwhelmed and talk myself out of it. It isn’t perfect. Every idea I bring to life now needs upkeep, and I have to tell it to stop agreeing with me.
What changed when I started using AI with ADHD?
I’ve always had a million ideas and no way to organize them. They’d pile up in notes apps, on sticky notes, and in my head at 2am, and almost none of them turned into anything.
Now I build tools left and right, some to organize my ideas and some to bring them to life. Working with AI has changed my life, and I don’t say that lightly, because I’m the person who rolls her eyes at “this changed my life” in a product review.
What changed is the part between having an idea and doing anything with it.
Why does AI understand what I’m trying to say?
My brain runs too fast, and when I’m talking to a real person the words don’t come. I know what I mean, and I can feel the whole shape of it, but by the time it reaches my mouth it’s in pieces.
With AI, I can word vomit. I type in a messy ramble with half-finished thoughts and no order at all, and it seems to know where I’m going with it. Then we get things done.
A TikTok from @superaudhdman put words to it. He says neurodivergent people love AI because you can hand it your thoughts and ideas, and it can spit them back out in a way other people can understand. The way I heard it, he was talking about how much it helps, and I saved it because it described me.
I can’t tell you whether that’s true for everyone with ADHD. I have no idea, and it might not be true across the board. It happens to be true for me.
How does AI turn a huge idea into a first version?
I think too big and too far into the future. So I hand AI the entire picture, every feature I’ll want in five years included.
It does something I never managed on my own. It tells me to slow down, picks what goes in version one, and puts all the other things on a list for later.
The most recent one was today. I started a holiday gift tracker, because we’re a last-minute mess most years, and I made it way too big. I wanted every feature I could think of. AI came back with “let’s start with this,” and things like holiday food prep and letting my sister in got moved to later.
That small version is the part I needed. Now I can play around with it and see what’s needed and what isn’t, instead of guessing all of it up front.
Before AI, I’d skip that step and go straight to overwhelmed. I wrote about that pattern in The Seed You Never Plant Can Never Grow, and it’s the same idea, except now I have help shrinking the seed down to something I’ll plant.
| What I bring | What I get back | What’s still mine |
|---|---|---|
| A long, messy ramble with the whole five-year picture | A first version I can start today | Deciding if the idea is worth doing |
| Half-finished thoughts in no order | The same thoughts, organized | The words, and the final call |
| An idea I love | A list of what can wait | Noticing when it’s agreeing too much |
What’s the downside of bringing every idea to life?
Now I can make all my ideas real, which is wonderful, and now I have so many of them. Every one needs upkeep and maintenance, so I’ve got all these little projects to keep running all the time.
When the problem was ideas that never got started, upkeep wasn’t on my radar. I ran into the same thing with the chores that came with things I wanted, in Why Do I Resent the Chores That Come With the Things I Wanted?. I wrote about the automation version of this on my building blog too, in I Automated My Way Into Overwhelm.
It also feeds the habit I already have of starting something new instead of finishing what’s in front of me, which I wrote about in Why Do I Start Things and Never Finish Them?. A new tool is one ramble away now, so I have to be more honest with myself about which ideas get a first version.
How do I stop AI from agreeing with everything?
I don’t want a yes man. I’ve asked AI to push back on me many times, and I have a devil’s advocate setup whose whole job is to poke holes in my ideas.
I don’t know if it always happens, though. Sometimes I notice it’s being agreeable and I say it outright: “Hey, don’t agree with me. Tell me why I’m wrong. Tell me why this sucks.” It answers differently when I ask like that, and some of my ideas don’t survive it.
Do I use AI to talk to real people?
Sometimes. When my thoughts are jumbled and I need to explain something clearly to a person, I’ll have AI help me organize it. It’s usually for work, not my husband or my family.
It’s always my own words. I don’t have it make something up for me. It puts my thoughts in an order someone else can follow, and I say what I meant to say.
Today’s action
Try this with one idea you’ve been carrying
- Pick the idea you keep thinking about and never start, the one that got too big.
- Type the whole thing out, messy and out of order, with every feature and every someday.
- Ask: “What’s the smallest first version I could start today? What can wait?”
- Then ask: “Now tell me why this idea is a bad one.”
- If it still looks good after both answers, start the small version this week.
Frequently asked questions
Can AI help with ADHD?
It helps me with two ADHD-shaped problems, organizing a messy flood of thoughts and shrinking a huge idea into a first step. I can’t speak for everyone with ADHD, and it doesn’t replace anything a doctor or therapist does. For me it closed the gap between having an idea and starting it.
Is AI neurodivergent?
I don’t know, and I’m not the person to answer it. I can only tell you what it does for me. It follows my rambling without needing me to explain things in order, and that’s the part that matters to me.
Is it cheating to use AI to explain what I mean?
I don’t think so, because the words and the ideas are mine. AI puts them in an order someone else can follow. If it starts making things up that I didn’t say, I don’t use it.
How do I stop AI from just agreeing with me?
Ask it to disagree, out loud, in plain words. “Tell me why I’m wrong” or “tell me why this sucks” gets a different answer than “what do you think?” I also have a devil’s advocate setup for bigger decisions, and I still have to notice when it slides back into agreeing.
What do I ask AI when an idea feels too big?
Ask for the smallest first version you could start today and a list of what can wait. Then build or try only the first version and see what’s needed before adding anything else.
Where I am with this
Using AI with ADHD helps me most when I hand it a messy, too-big idea and ask for the smallest first version, and it helps least when I let every idea become one more thing to maintain. I’m still figuring out which ideas deserve a first version. Pick one of yours this week and see what the small version looks like.
A book I made
The Lightest One
An undated ADHD daily book that never asks you to prioritize. Dump everything in your head, mark each thing light, medium or heavy, and do the lightest one.
Get the paperback →I run an affiliate program on the tools I build. Approved affiliates earn 20% on what their referrals pay, for up to a year. US only for now. See the program →
