What is AI Brain + Recommendations [Friday Wrap-Up]

AI brain is when you’ve come to rely on AI for even the most basic things — and after a week deep in Claude Max, I felt it creep back in. I’m sharing the three warning signs I’ve identified so you can catch it early, plus how bad sleep and brain fog made it worse.

I’ve also got recommended reading from Mike Schmitz on using Claude to script YouTube videos (not my approach, but a thoughtful one), and a More Perfect Union video that exposes Polymarket’s prediction markets for what they really are.

On my mind

  • AI Brian
  • Good Sleep
  • How much are we really on our phones?

Recommended Reading

Recommended Media

View the episode transcript

  • (00:00) – – What’s on my mind
  • (08:03) – – Recommended reading
  • (12:14) – – Recommended media

————
Streamlined Solopreneur is the podcast for solopreneurs who want to automate their business and take time off worry-free. Each week, Joe Casabona shares practical systems, tools, and strategies to help you reclaim your time and run your business without sacrificing your the rest of your life, or your health. 

Start with the free Solopreneur Sweep — a step-by-step method for finding where your business is losing time: https://streamlined.fm/sweep

If this episode helped you, leaving a review on Apple Podcasts helps other solopreneurs find the show — it only takes a minute and means a lot.

Connect with Joe on LinkedIn: https://www.linkedin.com/in/jcasabona/

Transcript
00:00:00.220 –> 00:00:09.240

Hey everybody and welcome to the Friday wrap up for April 17th, 2026 on Streamline Solopreneur.

00:00:09.460 –> 00:00:11.700

This is a short episode where I talk about three things.

00:00:12.200 –> 00:00:16.780

What's on my mind this week, recommended reading, and recommended media.

00:00:17.500 –> 00:00:22.340

This show is to help you automate your business so you can take time off for free.

00:00:22.960 –> 00:00:26.060

And hopefully this curation will help you think more about your systems.

00:00:26.800 –> 00:00:29.980

I'm your host, Joe Casabona, and here's what's on my business.

00:00:30.000 –> 00:00:37.120

mind this week. The first thing is something that my friend Bianca and I lovingly call

00:00:37.420 –> 00:00:43.740

AI brain. I'm writing a much longer piece on this, but I've been trying out Claude Code as I

00:00:43.960 –> 00:00:51.140

or Claude Max, Claude co-work, just a bigger plan for Claude. And I've been thinking about this

00:00:51.320 –> 00:00:57.600

because in the middle of the week, I got hit with this weird brain fog that felt familiar

00:00:57.600 –> 00:01:06.580

like I had experienced it before, and in fact I had experienced it before, in June and July of

00:01:07.250 –> 00:01:17.460

2025, where I realized that I was relying on AI for too much. Now, I recognized the warning

00:01:17.640 –> 00:01:25.180

signs very early, but it caught me differently this time because I wasn't using it to create

00:01:25.200 –> 00:01:32.160

content or even come up with episode ideas. I was using it to crunch a bunch of data and then help

00:01:32.240 –> 00:01:40.380

me draw conclusions, but I still felt like I had spent a lot of the last week talking to a robot

00:01:41.440 –> 00:01:49.380

and treating it like a person, which is something I try very hard not to do. So this is something

00:01:49.480 –> 00:01:54.940

that's on my mind. Again, I'm working on a bigger piece for this. Uh,

00:01:55.180 –> 00:02:02.400

where I mentioned the three signs that you might be getting AI brain and how to prevent it.

00:02:02.520 –> 00:02:05.780

And I'll just share the three signs with you because those are the things that I've worked out.

00:02:06.660 –> 00:02:11.920

The first is that you can't think about something for more than five minutes before running to an LLM for advice.

00:02:12.120 –> 00:02:18.460

And this is the thing that made me realize it is I was sitting there kind of stuck between multiple tasks.

00:02:19.500 –> 00:02:24.960

I was up in the dining room, which is just a terrible place for me to work because it's right next to the playroom.

00:02:25.599 –> 00:02:30.160

and it's, you know, it's between the playroom and the living room. And so there's just a lot going on.

00:02:30.200 –> 00:02:35.760

It's very hard to focus. And instead of deciding I was going to move, I was like, oh, I'll just ask AI.

00:02:35.940 –> 00:02:41.860

And then I thought, wait, I spent no time thinking about this. So if you can't think about something for

00:02:41.960 –> 00:02:46.240

more than five minutes before running to an LLM, that's the first and most significant sign, I think.

00:02:46.880 –> 00:02:53.280

The second is that you're not talking to the people who matter, people being the emphasis here.

00:02:53.380 –> 00:02:59.280

If you are deciding not to talk to potential customers, clients, or people in your audience,

00:02:59.390 –> 00:03:05.940

and instead just asking AI what it thinks about the avatar it has for your customers, clients,

00:03:06.190 –> 00:03:09.140

an audience, you probably have AI brain.

00:03:09.680 –> 00:03:14.920

And if you have no skin in the game, and this was the second thing that I noticed from this week,

00:03:15.040 –> 00:03:20.900

is, you know, I asked it to crunch a bunch of data and then look at a bunch of client calls

00:03:21.400 –> 00:03:23.100

and come up with common objections.

00:03:23.980 –> 00:03:25.180

And I, of course, compared that

00:03:25.260 –> 00:03:26.940

against my own anecdotal evidence.

00:03:27.700 –> 00:03:30.220

And I thought, all right, well, help me come up

00:03:30.260 –> 00:03:31.280

with a content strategy.

00:03:31.500 –> 00:03:34.580

Don't think of the actual content titles,

00:03:34.700 –> 00:03:36.720

just kind of like the pillars for the content.

00:03:37.660 –> 00:03:39.680

And then as I'm sitting down to plan my podcast,

00:03:40.600 –> 00:03:43.980

I thought, I don't care about this plan.

00:03:44.320 –> 00:03:45.740

I had no skin in the game

00:03:45.860 –> 00:03:48.840

because I didn't spend appreciable time thinking about it.

00:03:48.920 –> 00:03:56.400

And so I was able to course correct a lot quickly than I was able to in summer 2025.

00:03:57.540 –> 00:04:04.340

But it is, it's a something that requires to quote Matt I. Moody, constant vigilance, I think,

00:04:04.520 –> 00:04:12.500

something that you need to think about. Because what Claude Max especially, like what AI can do,

00:04:13.560 –> 00:04:18.820

what large language models can do with bigger context when,

00:04:18.900 –> 00:04:26.800

Windows truly seems like magic sometimes until it doesn't. And it might be a while before it feels

00:04:26.980 –> 00:04:31.280

like it doesn't and that thing snaps you out of it. And so you know, I think that you want to,

00:04:32.500 –> 00:04:38.880

I have a, I'm giving you a problem without a solution right now because the solution is

00:04:40.080 –> 00:04:47.360

basically just an outline, but, you know, I'll talk about it more on this show, how to prevent

00:04:47.580 –> 00:04:52.840

AI brain or at least the things that I'm doing to prevent AI brain.

00:04:54.440 –> 00:04:58.320

The second thing that is on my mind is getting good sleep. I've been getting bad sleep this

00:04:58.440 –> 00:05:02.860

week. I have been inexplicably, I shouldn't say inexplicably. I've been eating kind of like

00:05:02.940 –> 00:05:11.080

garbage too close to bedtime. And I've been waking up somewhere between 2.30 and 4 every

00:05:11.960 –> 00:05:17.340

morning. And it hasn't been great. Last night, I ate reasonably

00:05:20.020 –> 00:05:21.620

a reasonable amount of time before bed.

00:05:22.760 –> 00:05:27.680

And I slept all the way until 4.50, which, I mean, I usually wake up at 5 anyway.

00:05:29.040 –> 00:05:35.240

And so I just, I've just been thinking about maybe part of the reason where I felt

00:05:36.080 –> 00:05:43.800

AI brain or where I felt the need to lean on Claude more than I would have otherwise

00:05:43.820 –> 00:05:49.560

is because of the brain fog from not getting very good sleep.

00:05:49.780 –> 00:05:57.220

And that's also something to think about, like check in with yourself and try to get good sleep.

00:05:57.400 –> 00:06:00.600

You know, I'm not a sleep expert or sleep doctor.

00:06:01.680 –> 00:06:02.840

I'm a 40-year-old man.

00:06:03.340 –> 00:06:13.780

And as somebody who's now in his midlife, and I and someone who experienced a lot of sleep deprivation with three small kids.

00:06:15.320 –> 00:06:21.440

I know the importance of sleep. My brother said that one thing he was kind of jealous of me about

00:06:21.700 –> 00:06:28.800

is the fact that for a long time I did not need a lot of sleep to function well.

00:06:30.540 –> 00:06:35.200

And while that is true, right, I can usually get away with a day of not great sleep

00:06:35.680 –> 00:06:43.760

or I can one day after not having good sleep. I do see the difference between getting good sleep.

00:06:43.780 –> 00:06:44.740

and getting bad sleep.

00:06:45.780 –> 00:06:48.260

And you're definitely at your best when you get good sleep.

00:06:48.380 –> 00:06:50.660

So just something that's on my mind.

00:06:51.620 –> 00:06:55.820

A super secret bonus third thing that I don't have in my outline here,

00:06:55.920 –> 00:06:59.080

but that I've been thinking about a lot is,

00:06:59.800 –> 00:07:01.600

has to do with my latest LinkedIn post,

00:07:01.680 –> 00:07:06.160

which I will link in the show notes and the description.

00:07:07.280 –> 00:07:10.700

And that is just like being on our phones too much.

00:07:11.780 –> 00:07:13.200

I went to a coffee shop this morning.

00:07:13.400 –> 00:07:15.560

and literally everybody in front of me

00:07:16.000 –> 00:07:18.040

was staring down at their phone with the same face.

00:07:19.180 –> 00:07:20.560

And I was just looking around

00:07:20.860 –> 00:07:23.780

to make eye contact with another human being

00:07:24.740 –> 00:07:26.360

and I didn't.

00:07:27.200 –> 00:07:29.540

And this is coming off of the heels of me writing

00:07:31.220 –> 00:07:34.080

a pretty, I feel important piece

00:07:35.040 –> 00:07:36.840

for the Home and School Association,

00:07:38.020 –> 00:07:41.140

which is I'm president of my kids' home and school association

00:07:41.380 –> 00:07:43.120

and I talk about the importance of technology

00:07:43.200 –> 00:07:45.500

and waiting until our kids are 14 to give them smartphones.

00:07:47.860 –> 00:07:50.600

And this is something I feel increasingly strongly about.

00:07:50.900 –> 00:07:53.760

My year of digital detox is going well, maybe too well.

00:07:54.680 –> 00:07:58.040

Definitely not too well, but it's just something I'm constantly thinking about.

00:07:58.140 –> 00:07:59.560

And it's a little bit of a half-baked idea,

00:07:59.560 –> 00:08:00.920

but it's something I wanted to bring up here.

00:08:02.000 –> 00:08:02.200

Okay.

00:08:03.480 –> 00:08:07.880

So, recommended reading and media.

00:08:08.840 –> 00:08:14.100

Recommended reading is an article from my friend Mike Schmitz called How Clawed

00:08:14.260 –> 00:08:21.060

helped me make the videos I want to make. And he goes through this process about how he has

00:08:21.220 –> 00:08:27.320

started a new YouTube channel about basically book reviews. He has the bookworm podcast,

00:08:27.500 –> 00:08:37.460

which is really good. And he was stuck because he had a certain quality in his mind for

00:08:37.460 –> 00:08:45.860

these videos and scripting them seemed like something he'd be able to, like something he wouldn't

00:08:45.860 –> 00:08:55.200

be able to do consistently. And so he asked himself, I'm quoting the article here, could I expedite

00:08:55.200 –> 00:09:01.200

the process by hiring Claude as a research assistant and scriptwriter? And he says the answer is a

00:09:01.480 –> 00:09:07.420

resounding yes. And now he's personally walked me through this system because

00:09:08.500 –> 00:09:09.440

we're in a mastermind together

00:09:10.420 –> 00:09:10.820

and

00:09:12.140 –> 00:09:12.920

I am

00:09:13.660 –> 00:09:15.280

definitely the town crier

00:09:15.760 –> 00:09:17.300

for AI and overuse of AI

00:09:17.540 –> 00:09:18.440

in this mastermind group

00:09:19.620 –> 00:09:20.760

I am easily

00:09:22.100 –> 00:09:23.780

I would say I'm easily the most skeptical

00:09:23.900 –> 00:09:25.320

person I don't know if that's true

00:09:26.840 –> 00:09:28.620

but I feel like I am

00:09:28.620 –> 00:09:29.960

or I feel like I'm at least the most

00:09:30.300 –> 00:09:32.060

vocally skeptical about it

00:09:34.240 –> 00:09:35.480

but he walked me through this whole process

00:09:35.490 –> 00:09:36.220

and it was really interesting

00:09:36.400 –> 00:09:43.120

he takes really, really in-depth notes with like an emoji system and things like that.

00:09:43.120 –> 00:09:46.000

And he quote unquote trained Claude on this.

00:09:46.000 –> 00:09:48.120

He has a video script template.

00:09:50.100 –> 00:09:58.600

And so he feeds all of this into a Claude skill that pulls in all of his personal notes,

00:09:59.140 –> 00:10:02.060

pulls in the mind map, grabs his rating for the book,

00:10:03.460 –> 00:10:09.620

pulls in the bookworm episode, which is a full transcript to get his thoughts on the book,

00:10:09.800 –> 00:10:15.600

interprets his emoji coding system, and puts it together. And so this is really interesting

00:10:15.720 –> 00:10:24.120

because this is a, this is a riff on the question I get a lot, which is if I train AI on everything

00:10:24.280 –> 00:10:31.480

I've ever written, doesn't it make something that sounds like me? And I think what's really

00:10:31.500 –> 00:10:33.300

important here is that

00:10:34.280 –> 00:10:34.920

Claude is

00:10:36.100 –> 00:10:37.240

writing the video script,

00:10:37.440 –> 00:10:39.600

quote unquote, which I question

00:10:40.520 –> 00:10:41.580

but I also

00:10:41.700 –> 00:10:43.440

watched Mike's video and it felt

00:10:43.540 –> 00:10:44.680

very natural, so I don't know

00:10:45.760 –> 00:10:47.620

I don't know what the delta is between

00:10:48.280 –> 00:10:49.720

the thing that Claude outputs

00:10:49.860 –> 00:10:51.460

and then the thing that he modifies.

00:10:52.780 –> 00:10:53.020

But

00:10:54.540 –> 00:10:55.620

the AI is also

00:10:55.760 –> 00:10:57.140

not the delivery mechanism.

00:10:57.600 –> 00:10:59.140

So he takes this script,

00:10:59.320 –> 00:11:01.460

probably fixes it up. The script

00:11:01.560 –> 00:11:08.300

is heavily influenced on a very narrow topic, I should say, right? He's not saying come up with a book

00:11:08.500 –> 00:11:17.520

review for me. He's saying here is a lot of assets about a book I've read now put together a script

00:11:17.520 –> 00:11:26.380

in this template for me. And then he delivers it. And so I think that this is an interest,

00:11:26.480 –> 00:11:31.160

I'm going to say an interesting approach. It's not for me.

00:11:32.400 –> 00:11:33.200

I don't think

00:11:34.580 –> 00:11:34.700

but

00:11:35.640 –> 00:11:35.860

you know

00:11:36.040 –> 00:11:37.500

he records the video

00:11:37.920 –> 00:11:39.560

and then he ships it off to an editor

00:11:39.820 –> 00:11:40.420

who edits it

00:11:41.940 –> 00:11:43.140

and so his bottom line

00:11:43.260 –> 00:11:44.680

is Claude helps him make videos

00:11:44.730 –> 00:11:46.420

he wouldn't otherwise be able to make

00:11:47.900 –> 00:11:48.340

so

00:11:50.220 –> 00:11:50.880

I'm you know I

00:11:51.920 –> 00:11:53.160

again I'm going to say this is

00:11:53.380 –> 00:11:55.140

not a process for me but I think

00:11:55.260 –> 00:11:57.180

it's an interesting process and I don't want to have

00:11:57.240 –> 00:11:59.060

like some monoculture

00:11:59.240 –> 00:12:01.160

on this podcast I want to

00:12:01.640 –> 00:12:07.960

present other points of view, even if I do present mine the most because it's my show.

00:12:09.020 –> 00:12:12.860

So I think this is really interesting. I'll have a link to this and everything in the show notes.

00:12:14.000 –> 00:12:20.040

And then recommended media. I was going to recommend the Scrubs revival, which we'll call it season

00:12:20.200 –> 00:12:26.840

10 of Scrubs. One of my favorite show, maybe my favorite show of all time, usurping friends for

00:12:26.960 –> 00:12:31.120

that spot. And I thought that the nine episode revival,

00:12:32.579 –> 00:12:33.380

was fantastic.

00:12:34.740 –> 00:12:35.300

And so I love that.

00:12:35.380 –> 00:12:36.100

I love that show.

00:12:36.400 –> 00:12:37.740

I hope it comes back for season two.

00:12:38.400 –> 00:12:39.840

They opened up a lot of threads.

00:12:40.860 –> 00:12:42.660

And so I'd imagine they're pretty confident

00:12:42.820 –> 00:12:44.160

it's going to come back for a season two,

00:12:44.280 –> 00:12:46.020

but I don't think it's been officially renewed

00:12:46.580 –> 00:12:47.900

as of the time of this recording.

00:12:48.820 –> 00:12:51.000

So that was my original media recommendation

00:12:51.360 –> 00:12:53.720

until I watched a video from More Perfect Union

00:12:54.980 –> 00:12:57.520

called Polymarket Asked to Work with Us.

00:12:58.120 –> 00:13:00.180

We exposed their scam instead.

00:13:00.220 –> 00:13:08.200

It's a 20-minute video that I think does a, I have no opinions on the outlet more perfect union.

00:13:09.260 –> 00:13:17.760

I have not consumed a lot of their content, but I certainly liked this.

00:13:18.760 –> 00:13:29.540

And so I have been suspicious of things like this, Polymarket, and the other one they mentioned,

00:13:29.900 –> 00:13:32.880

I don't think it's called Kashi.

00:13:34.740 –> 00:13:36.400

But whatever it's, uh,

00:13:37.140 –> 00:13:39.200

Kalshi. Close. Really close.

00:13:39.320 –> 00:13:39.840

Call she.

00:13:40.270 –> 00:13:44.580

Uh, about, you know, democratizing financing and,

00:13:44.760 –> 00:13:46.480

and really just a bunch of scams, right?

00:13:46.620 –> 00:13:51.540

Anybody who fights this hard to say they're not a scam almost certainly is a scam.

00:13:52.580 –> 00:13:56.200

This is based, and it's gambling and they go right of their way.

00:13:56.280 –> 00:13:57.880

It's really hard to say it's not gambling.

00:13:59.500 –> 00:14:09.440

And I just think it's a very well-done video that looks at, you know, that if you've never heard of polymarkets or in sports, this is called prop betting.

00:14:09.600 –> 00:14:15.440

And prop betting is bad, really bad, really bad.

00:14:16.520 –> 00:14:19.800

Death threats to players bad, right?

00:14:20.220 –> 00:14:29.220

Players throwing, like doing the thing that Pete Rose has long been accused of or like the, the Black Sox scandal.

00:14:29.640 –> 00:14:34.920

of 1919 on a much more granular level.

00:14:37.760 –> 00:14:49.080

So it's just like a combination of a really low barrier way to gamble and a really easy way

00:14:49.080 –> 00:14:52.060

to do insider trading if you're the insider.

00:14:53.880 –> 00:14:55.480

So I thought that was a really good video.

00:14:55.490 –> 00:14:57.360

I think that if you don't know anything about.

00:14:59.160 –> 00:15:27.960

Polly Market, which is there are these prediction markets. And so you can bet on like what, how many days this ceasefire will be or how many pitches Max Fried will throw in his next outing or like, will Trevor Noah say potato? That's like right from the video. Will Trevor Noah say the word potato when he hosts whatever he's hosting?

00:15:29.180 –> 00:15:32.560

And so like a lot of knowable things for the right people.

00:15:33.180 –> 00:15:34.040

So I think it's really good.

00:15:34.040 –> 00:15:34.780

I was strongly recommended.

00:15:35.600 –> 00:15:37.060

So that's my media recommendation.

00:15:37.420 –> 00:15:44.200

So that's what's on my mind, some recommended reading, and some recommended media.

00:15:44.920 –> 00:15:51.300

That's it for this episode of the Friday wrapup at Streamline Solopreneur.

00:15:52.780 –> 00:15:58.760

If you have thoughts or a story, you can send it over at Streamline.

00:15:59.840 –> 00:16:00.800

Feedback.com.

00:16:01.840 –> 00:16:08.180

If you want to think more about your systems as a solopreneur, maybe you're interested in

00:16:08.280 –> 00:16:15.120

trying something like Mike mentioned here, then you can take a closer look at everything you

00:16:15.220 –> 00:16:15.420

do.

00:16:15.820 –> 00:16:21.200

And I have a process for how to do that over at streamlined.fm slash sweep.

00:16:22.200 –> 00:16:23.280

That's my solopreneur sweep.

00:16:23.500 –> 00:16:24.420

It's a free process.

00:16:25.080 –> 00:16:27.500

It's like a 30 minute audit of your business.

00:16:28.560 –> 00:16:32.240

So I would check it out, maybe ask yourself those questions.

00:16:32.540 –> 00:16:35.840

Again, that's over at streamlined.fm slash sweep.

00:16:36.540 –> 00:16:37.520

Thanks so much for listening.

00:16:38.300 –> 00:16:42.660

And until next time, I hope you find some space in your weekend.

Join the Discussion