Using AI to think with you when your focus is unreliable
I had the kind of day people ask for. Two short meetings, and the rest of it mine.
I spent the first thirty minutes of it staring at my screen.
Not thinking. Not prioritizing. Staring. I had real deadlines and a follow-up list longer than the week could hold. None of it moved me. My task initiation, executive function, whatever you want to call it, was not working at its optimal level that morning, and knowing what was at stake did not change it.
I knew what needed to be done. Knowing did not help me start.
This is not a rare morning for me. Since menopause, my focus and short-term memory are not reliable every day and I have stopped treating that as a scheduling accident.
Why the brain dump wasn’t enough
Long before I had any AI tool, I had the brain dump. Open a note, get everything out of my head, pick something.
Sometimes it worked. Sometimes it did not, and it took me a long time to see why. I was still the engine for that method. Something in me had to start it, and there was nothing external to prompt me or wake my brain up and get it engaged. On a morning when starting is the thing that isn’t available, a method that requires you to start is not a method. It is one more item on the list.
The usual advice did not solve that problem. Breaking the work into smaller steps still required me to begin.
What got me moving
I needed something outside me to ask the first question.
So I did the only thing I could think of, which was talk to my AI tool, using the microphone rather than typing.
I unloaded. Literally unloaded. What I was responsible for, what was stressing me out, what was expected of me, the personal appointments in the way, how little usable time I actually had.
And to my surprise, it didn’t give me a plan right out of the gate. It asked me five questions instead.
Which items have a real deadline today, and what happens if they do not get done?
What is already on the calendar, and how much usable time is actually left?
Is anyone blocked waiting on me?
What is my energy and focus capacity today?
If I completed only one meaningful thing, what would create the most relief or progress?
Answering took about sixty seconds. Somewhere in the middle of it, something woke up in my brain. I could see what was important and what was actionable.
Once I answered the questions, the tool gave me a plan, in two versions. One for the day I had described, and one in case my capacity dropped. Lower-energy, less demanding, still real work. I had never thought to ask for that.
Here is the honest part. I do not think the plan itself was novel. What the tool really gave me was a better question. I had been used to asking how far I could push myself. It handed me a different one: what is my work ceiling today. The shift was practical. I stopped asking how far I could push and asked what I could reasonably carry that day. That took enough pressure off for me to work steadily through high-priority work for the rest of the day.
That lower-energy version never went away. It is now a permanent setting in the way I track my work, with a saved view I open on the days I need it. I still use that Low-energy view.
Where it holds now
I wrote recently about a conference where an earlier practice of mine broke. I could not preserve what mattered during a day of back-to-back sessions, and some follow-ups never went out.
I have been to a few conferences since, and it does not break there anymore. The difference is that I capture first and think later.
At the start of each break, not the end, I step somewhere quiet and talk for no more than two minutes. Whatever I can still reach, out loud, into my phone. Four of those got me through the last conference.
Two minutes at the start of the break, before the next session takes the room.
The thinking happens afterwards, when I have capacity again. I open up my AI tool and say: I do not remember a thing that happened at that conference, I need to identify my key follow-ups and who I need to talk to, how can you help me. The tool is my memory so that my brain can be at full capacity for the conference itself.
That changed the result. I could be present for the next session, and every follow-up went out after the conference.
The tradeoff
The tradeoff is that AI can take more time than it saves.
AI keeps offering more: more context, more improvements, more adjacent ideas. When I need one plan for today, more becomes a distraction.
I would get stuck in idea mode and neglect execution mode, and it felt productive the whole time, which is what makes it expensive. While laying the groundwork for this business, I followed a tangent far enough that I spent two solid weeks building process documentation, workflows, and infrastructure playbooks. All of it pristine. None of it moved the business forward.
What fixed it was not discipline. It was giving my AI tool the outcome I want and the constraints up front, so there is a stopping point before it starts.
The judgment never goes over. I decide what matters, what gets done, who I talk to and how. I ask for sources when something is stated as fact. I ask how it arrived at that, explain your rationale, explain it like I am ten. Then I read all of it and decide what holds.
I still make the judgment calls. AI helps me shape what I surfaced so I can act on it.
Where to start
You do not need my setup. Mine is connected to the system I run my work in, which helps, but that is a refinement, not the price of entry.
The next morning nothing starts, open a blank chat and say this out loud:
I am stuck. Here is where I am at. Here is where I would like to be. How can you help me?
Then tell it what you want out of the conversation and where to stop. Ask for one bounded thing you can act on today.
If you want the fuller version of the move, the 4 Moves Starter Guide walks through Surface, Shape, Practice and Tune, with a worksheet you can use today.
