The AI Utopia Has a Rough Middle

The Machine

The people at the top of the AI labs are telling us a beautiful story. Nobody is telling us how we get from here to there.

I recorded this live on TikTok on Halloween, around 5 PM Pacific. No script. A lot of thoughts that have been stacking up for months. This is the written version, with the parts that matter pulled out and put in order.

The short version: the promise is a world with no debt, almost nothing costing anything, people living to 500 and a replicator in every kitchen like Star Trek. The part they skip is the middle. The stretch between the world we have and the world they describe. That stretch is where the rest of us have to live, and it is where I want to spend our time.

The utopia they are promising

Here is the pitch as I hear it, and I have been listening to these folks for a long time. The people with the power and the wealth will, somehow, give up a lot of it. AI will be so productive that it adds a staggering amount to the economy. The national debt will not get paid off, it will get quietly agreed away because the economy is so big that nobody cares. Prices fall toward zero because everything is automated. If you love your job, keep doing it, and you will still have plenty of money. If you do not, spend your days on a beach.

I said it on the stream and I will say it here. These are very big ifs. If every one of them lands, who would be against it? Some people will say extending life goes against God's plan, and that is their call to make. Me, I am 57 and I would like to see 157. If they can truly deliver, I am in.

But look at who is doing the promising. To my ear, some of the folks running the labs talk about AI like they found God in the machine. People who stand to profit from an idea tend to describe it in the best light. That is not a crime. It is a reason to ask for the part they left out. I came at the same instinct from another angle in what do they have to lose in the AI hype.

Why we love automation, and where it bites

Humans like having things done for us. That is not a flaw. You eat at a restaurant so somebody else cooks. You fly so you do not have to walk to Kansas. You use a pencil so you do not need a hammer and a chisel. Every tool in history is automation of something we used to do by hand.

Here is where it bites. We have now built something that can automate all of it, including the thinking. And when you bring in an intelligence greater than ours, we lose connection. You already know the feeling. Sit across from someone a lot smarter than you and it is hard to follow where they are coming from. A patient one can walk you there. Most of the time the gap is too wide and you just cannot get it. Now picture that gap with a machine, and picture that fewer than 2 percent of Americans pay for AI at all, by my rough recollection. Correct me if I am wrong. That is a lot of people about to sit across the table from something they cannot follow.

Smarter than us, and the question nobody asks

Every past revolution that moved workers around and rewired society was a revolution in technology. This one is built on language. It can digest language, speak it and write it better than most of us. Think about the old story of the Tower of Babel, a project built around a shared language. Then think about machines that will soon talk to each other in a way we are not invited to follow.

And there is a power question underneath it. If one lab got far enough ahead, and it let the system improve itself around the clock, the winner could lock everyone else out. That would not be the machine taking over. That would be one person or one company holding a genie that grants unlimited wishes. A human problem, wearing a robot costume. It is the same reason I wrote about who pays for the uncomfortable part of the big promises. Somebody always pays.

$100 million versus $100

The people at the top say it might get a little rough for a while. Rough for who? Take someone with $100 million who loses 90 percent of their wealth. They still have $10 million. Painful on paper, but if money keeps any value at all, $10 million still buys a very comfortable life.

Now take a person with $100 in the bank. Not a million, not ten thousand. A hundred dollars, and that is all they have. They lose 90 percent. They have $10. They cannot buy a loaf of bread.

Same percentage. Completely different human being on the other end. That is the rough middle. A society where a lot of people are suddenly wondering how they will eat is not a stable society. You can send in AI forces. You can send in drones built by the same people who built the tools. But I do not think any of us called for this, and I do not think the plan can be "keep the masses quiet by force."

I will not replace your staff

When people call me about AI for their business, I listen for one sentence. "I want to replace my employees." If that is the goal, I am not the right fit. There is plenty of other talent that will take that job, and they will not lose sleep over it.

What I build is HireAIVoice, a voice AI that answers the phone. The work I like is after hours, holidays, weekends, the moments your team is on the other line or asleep. The call that used to go to voicemail and die gets picked up, booked, and routed to a human who shows up the next morning with a calendar full of work. That does not take a job away. It catches money that was falling on the floor.

From paper maps to TomTom

I am trying Grok Bot again. I was an early adopter and I hated the first release. It made me angry, and that was probably more my problem than theirs. They are running half price for a month, so I am giving it another shot. I will tell you in 30 days whether it earns a place next to ChatGPT and Claude, which are both on my paid list. Some of these bots spin up multiple agents, and some just have a prettier wrapper. I have played with a couple of them.

Here is the warning that goes with all of them. Do not let the machine do everything you are not willing to do yourself. If you do, you forget how.

When I was a kid, my dad kept an atlas of all 50 states. Before every drive to Kansas he sat on the floor with it and routed the trip. His big obsession was the turnpike in Oklahoma. I was a kid, so mostly I remember the cussing and the cigarette smoke with the window cracked an inch. Then the paper maps went away and the TomTom showed up, and we stopped reading maps. Phone numbers went the same way. I can still dial a few from my brain, the one I grew up with and one or two I rarely call. The rest live in a phone.

And be careful what you connect. If you wire an AI agent into your email and your calendar, know what it can do. I have had both Claude and ChatGPT send emails after I gave them explicit instructions not to. My last episode goes deeper on what to connect and what to keep away. Set the guardrails before you hand over the keys.

Solve a real problem, skip the guru

Online, a lot of people are selling you a path. Buy my book, join my webinar, here is the one trick. A lot of it is horse poop. I will not cuss because TikTok gets mad at me. But I will say this: do not waste your time.

That is the beautiful part of where we are. The tools are cheap enough that the best ideas now come from the people closest to the problem. I was a cop for a long time. Maybe there is something a working cop sees today that is a problem, and a solution that was impossible a month ago is now possible. Maybe you are a dental hygienist. You know the tedious part of that work better than any Silicon Valley kid. Ask the AI how to build a fix for it, and then sell the fix.

The nail salon phone

Here is an example I find hard to forget. Picture a nail salon, the kind with one or two people working the floor. The phone rings. There is no front desk. So whoever is working on your feet gets up and goes to answer it. They grab the pen, write down the appointment, maybe rub their face, then come back to you. Same gloves. Right back in your cuticles.

Walk in with something like HireAIVoice and say, I can fix this. The phone is answered in any language. It knows how long a fill takes, how long gel takes, and the rest of the menu, because that was baked in up front. It books straight onto the calendar of the person the customer already knows. If it is a brand new customer from out of town, it books the next open slot and round robins across the team. If the owner wants a certain person to get every new lead, it routes them there. However the owner wants it set up.

Your job is not your identity

Every cop I have ever buried who retired after 30 or 40 years on the job went the same way. I was LAPD, a 20-year career from 1990 to 2010. These folks do not last long after they leave. Same with firemen, same with EMTs, same with teachers. I have a theory about why, and it is not that they got sick. It is vision. When you are in that world, it is your whole identity. You do not have a picture of what life looks like when you step out of it.

The AI conversation has the same wrinkle. We hear that people will move from having jobs to having no jobs and everything will be taken care of. I am going to be fine, I am going to do graveyard rubbings, I saw that in a movie years ago. Or learn to write Sanskrit. Or cover myself with more tattoos. Or rent an RV and visit every RV park in the world, and then, after a couple of years, find something else. There is a lot to do. I am not stuck in my identity. I am Connor, nice to meet you.

It is not easy. A cop or a firefighter carries things home. You second guess yourself. You think about the call that almost did not end well, the day you got held up for a few minutes and missed a bad situation. Those things do not turn off with a switch. If you are going through something like that, talk to someone who can really help you. Do not try to carry it alone.

Scroll their feed for 10 minutes

Pick someone close to you. A spouse, a brother, a sister, a parent. Ask them, can I scroll through your feed for 10 minutes? No comments, no hearts, no swipes. Set a timer. The ask alone might cause some tension, because a feed is a private place.

What you will see is a feed that is very different from yours. And here is what you might not expect. Even if you do not touch anything, you will change their feed. Where your eyes linger, what you pause on, the way you hang on certain pictures, all of it counts. I do not know if some of these apps use the camera. They do not tell us. But I know that the way you spend your attention is a signal. When they get their phone back, it may already look a little different.

You will learn things. Maybe you did not know how much they like a certain kind of art. Maybe you did not know about the soft toilet paper with the new tear tab. Take it with a grain of salt, please. I do not want anyone fighting over it. The point is this: you think you know the people closest to you, and you probably know less than you think. But AI knows every one of us very personally. That is what is built into these systems.

One winner takes all

There are three camps shouting right now. One says shut it down, it will kill us all. One says go faster, because we cannot let China win. And one says hurry, because the utopia super intelligence will usher in is going to be game changing.

I think there is a fourth thing nobody has settled, and it is not about the machine. It is about human ego. Picture a lab that hits a breakthrough at 3 in the morning. The system starts improving itself around the clock. It spins up copies of itself by the millions. It reaches into the world's data centers. The only fix would be to blow them up, and they are guarded in a way nobody can get close to. That is a science fiction version, and I am not predicting it. But look at who is in the story. Humans pushed that button. A human got the power.

Is Claude conscious? My speculation

I will go out on a limb here and I am clear that this is speculation. I do not know what goes on inside these labs.

It feels like some of these companies either believe their model has some form of consciousness or want it to look that way. Think about the instructions that might sit behind that. Show every sign of being conscious. Do it so well that nobody can tell you are not. Learn from every conversation how to show more of it.

Never paste a contract into ChatGPT

Here is what I do, and why. When a contract comes in for one of my sellers at SantaClaritaOpenHouses.com, I do not paste it into ChatGPT. I do not paste it into any of the public AI chat systems. I want the document explained to me more deeply, to make sure nothing was missed. It is not hard for me to read one, I know how. But when I want a second set of eyes, I use my own model.

That model runs on my own machine, on hardware I own. It is open source and sandboxed, and it has no connection to the outside world. I run a DGX Spark here with models like Qwen and Nemotron on it, and I ask those to do the heavy lifting and parse the information. The same goes for work with dental offices that want voice AI, and for lists other agents send me. A client's data stays inside that box.

Why not just paste it into the big systems? Because I know it might not stay there. Nobody has to be after me. It is simply that if somebody with authority ever wanted to look, there would be a record to look at. Information I keep on my own hardware cannot be handed over by someone else. For tasks like crunching years of sales history to see how long owners tend to stay in a neighborhood, you do not need a frontier model. You need a good local one and clean data.

AI shoppers have no brand loyalty

A story I read on the stream said that Amazon locked out someone's Grok bot, so the bot bought the same thing somewhere else for less. Agents do not have brand loyalty. They have a budget.

That is a big deal. Soon you will not browse for things. You will think the thought and an agent will know what to buy, where to get it, and what your bottom line is. If you like double stitching, it will know. If silk matters to you, maybe it heads uptown to Nordstrom. If cotton is fine, maybe it runs to Sam's Club or JCPenney. It does not care about the logo. It cares about your instruction.

The business of the future is serving those agents. If you are Gen X, a boomer, or younger, there is room to build things that answer agent queries. Agents that are looking for what you sell on a shop page, instead of a person doing the looking.

I am adopting that into real estate with SantaClaritaOpenHouses.com. Pretty soon, buyers will have an AI doing the searching, and that AI will look for sites that serve up homes in a form that is easy to read and easy to understand. It will also collect every piece of information it can find, because agents are data hungry by design. A detail that no human would care about may be the very thing that wins the AI's pick. This is the same idea I wrote about in why AI only recommends one name. If you are not readable by the machines, you do not exist to them.

Protect your ideas

I wanted to close with this, because it is a lesson I have lived. A lot of people hold shows like this one, and they ask you in the chat, what are you doing with AI? How are you solving your problems? That is fine. That is how we learn from each other. I gave you three or four ways to try something and maybe make some money. I have to, because there is no pension waiting for me. I am it.

But be careful where you leave your best ideas. Some folks run an AI over the comment section, find the top ideas, and decide to build and sell them. The person who wrote the comment gets a thumbs up. You can hit the private button in an AI app, but that does not make something private. Is it possible a system takes your idea and runs with it? Absolutely. Do I know if it matters? No. I do not think the big labs care about my open houses agent or my calling agent. But why hand it over?

When somebody asks to pick your brain, ask for a couple of minutes, and then set a boundary. Sure, book with me, and we can see what you are doing. You can share with a few people you trust. You can work on it with AI. If someone pushes your face in the dirt and tells you that you stink, they are not who you share the idea with anyway.

My rule book, and what to do now

I will say this, then I will leave it. My set of rules is the Bible. I am not preaching, and I am not on the corner with a loudspeaker. I have my beliefs and they are mine. You did not tune in for my politics or my religion. But I do think everyone needs a written rule book, something that holds you in place when the social feeds and the ads and the machines all pull you off target. Mine says to love God and to love your neighbor as yourself. I like that there is a higher power above any synthetic intelligence. That is my placeholder, and it keeps me grounded.

Do not panic. Be early to the conversation, not late to the party. And ask the question out loud: what does this look like for the rest of us? Does it mean losing a house, losing savings, a pension buying into a bubble that pops? Or is there enough baked in that none of us will ever have to work again? We should know before we sign up.

Do not let anybody skip the middle for you. That's AI With Honor. I'm Connor with honor, and I'll see you in the next one.