Private notes, made public. · 30th Oct, 2026

Hey, it’s Gabriel.

These are my private notes this week on building wealth, raising my family, and breaking generational patterns.

This week
⚡ Stop adding AI to your business
❤️ Why I always ask for a bite
💰 Priced for perfection
🧬 I just started moving more

Y Combinator, the accelerator that backed Airbnb and Stripe, gave founders a piece of advice I keep coming back to.

Don’t build AI tools and sell them to law firms. Build the AI-native law firm yourself.

Here’s why that’s right.

Legacy companies are legacy. They have too many moving parts and too much to lose. Changing how the whole company works is a huge risk, so they don’t. What they do instead is slap AI on top and keep every workflow exactly the same.

But the biggest leverage AI gives you was never about making something 30% faster. It’s about rethinking how the work gets done in the first place.

I learned this the painful way.

When I tried to bring AI into my own marketing agency, I was still planning around the people. Every plan started with “who does this step?” It was very hard to reinvent a workflow when the workflow was built around jobs people already had.

Then I started Two Hour CEO. I wanted to run it alone, as one man, as far as I could take it. That’s when I saw it.

Think about what it takes to make one video ad the old way:

And the whole cycle starts again.

It’s a relay race with six runners, and the baton gets dropped at every handover.

My first instinct, back at the agency, was to give each of those six people an AI. It made them faster. It was still six people and five handovers.

At Two Hour CEO, one AI agent runs the whole chain.

I give it an idea. It writes. Then it designs and generates the video, and uploads it straight into Meta through the API. Every day it reports back which ads worked and which didn’t. Then it optimises, makes new creatives, and runs the cycle again.

One agent.

It still needs someone to manage it. Right now that’s me. Eventually I’ll hire one person to run that agent end to end.

That’s the new world. We won’t hire specialists for one step anymore. We’ll hire people to own a whole value chain, start to finish.

The second thing I’m seeing is harder to say out loud.

A lot of companies are going to suffer. They can’t risk changing anything in case something breaks, so they’ll keep slapping AI on top of the old workflow.

The few that make it will do something uncomfortable. They’ll pull out a small team and have it rebuild the whole value chain from scratch, in parallel with the legacy team, chasing the same result side by side. Only when the new way wins do they move the legacy team somewhere else, or hand them the new chain to run end to end.

It’s a tough conversation. Most companies won’t have it.

So I think we’re going to watch a lot of legacy companies die in the months and years ahead.

And most of them won’t die suddenly. It will be a slow death, because changing is so hard. It’s the frog in the pot. The water heats up one degree at a time, and nothing ever feels urgent enough to jump. By the time they realise it, it will be too late.

If you started your company again tomorrow, from zero, how many people would it take to make one ad?

This is what we build inside Two Hour CEO. One person, running agents across a whole value chain instead of one step. Doors are closed right now, but the waitlist is open, and waitlist members hear first when we open again. Join the waitlist here.

I took my kids out for lunch this week. After lunch, they always want dessert. So I bought them ice cream, as usual.

And as usual, when I handed it over, I asked: “Can I have one bite?”

They gave me a bite. Then I left them alone.

This time I asked myself why I always do that.

The honest answer is that it’s a little test. Are they willing to share? Do they see the ice cream as theirs, or can they give some of it up?

Because the truth is, the ice cream doesn’t really belong to them. I paid for it.

That’s when it hit me. This is what tithing is.

The church gets a bad reputation for tithing, and a lot of it is earned. Many prosperity gospel churches manipulate people into giving. “Tithe, and God will give you more,” like He’s a slot machine. They make it a transaction. Or they condemn people and make them feel guilty for not giving.

I’m very blessed to be in a church that believes tithing has to come from revelation. Even if you don’t tithe, you’re still loved and accepted. Tithing is an act of worship. And you can only worship like that once you’ve had a revelation of how much God loves you.

When God asks us for 10%, He’s asking for a bite of the ice cream.

He doesn’t need it. He bought the whole thing. He’s asking whether we know that nothing in our hands is really ours, and whether we can open our hands and give back the 10% that was His in the first place.

God doesn’t need our tithe. But He wants us to tithe, as an act of worship and love.

And here’s the funny thing about His grace. When we give in faith like that, we’re always rewarded.

I’m not twisting His word into “I gave You something, so You owe me.” That’s the slot machine again. I give, and I trust that God will reward me in His time. In my own life, that has always been true.

What are you holding so tightly that you’ve forgotten who paid for it?

The Setup: I’m mostly holding right now, not adding. I still don’t believe we’re in a dot-com bubble. But I do think the market is due for a correction, and I’m being more careful.

The Data:

So why be careful? Because everything is priced for perfection.

The market always looks forward. A lot of the growth is already in the price. So even when a company beats earnings, the stock may not go up, because the beat was already expected. To go up, results have to be even bigger than the big numbers everyone’s waiting for. That’s a lot less likely.

Which means the upside is now small, and the downside is big. Miss by a little and the stock gets punished.

That’s exactly what happened this week.

The Example: I’ve been buying a little ServiceNow, as part of my earlier view on the SaaS apocalypse: some good software companies are getting dragged down with the bad ones. ServiceNow is growing and it’s profitable. On 28 October it reported earnings of $1.00 a share. Analysts expected $1.02. Revenue actually beat expectations.

It missed by 2 cents. The next day, the stock fell 27%. Its worst day ever.

The Honest Part: This is the market we’re living in. A great business can drop a quarter of its value over two cents. I’m not panicking, and I’m not adding much either.

Next issue, I’ll talk about Anthropic’s listing. Reports put it at $1.5 trillion or more, and I’ll share what I think it means for the AI world. (None of this is financial advice. It’s my own money and my own views. Please do your own due diligence.)

A lot of the time we go looking for the newest supplement, the coolest gadget, the hyperbaric oxygen chamber, the ice bath, whatever’s trending.

But most of it comes back to three simple things. Sleep. Movement. Diet.

That’s it. That’s the 80/20. The 20% that gives you 80% of the result.

This week I picked one of the three and decided to just move more, all day long.

Here’s what that looks like for me:

What changed: I can feel a huge boost in my motivation, and a huge boost in my mood. And that helps me work better.

It’s day and night compared with the last couple of weeks. I bought a new office chair recently. It’s very comfortable. Too comfortable. I sat in it all day, every day, and I felt it.

Movement is underrated. Not a workout you have to psych yourself up for. Just moving, all through the day.

How many hours did you spend sitting yesterday?

As promised, my private notes. No fluff.

See you next week,

Gabriel Judah

P.S. Inside Two Hour CEO, I show members how I run a whole company with one person and a team of AI agents, including the ad agent from this issue. Doors are closed for now. Join the waitlist and you’ll hear first when we open again. Join the waitlist here.