AI is a trojan horse. It looks like a gift. You let it in and hand over the thinking, and keep that up long enough and you lose the ability to do it at all. Pair that with today’s brain-rot, doom scrolling social media, and you’ve got Idiocracy in the making.
Let me explain.
What is AI, in simple terms?
It’s a fancy auto-complete. You give it an input, it matches that against its training data, and it spits out the most likely response. That’s it. The data is what feeds the answer.
Here’s the part people skip over. That training data isn’t all high-IQ research and textbooks. Some of it is. But a huge chunk is just the open internet. Websites. Forums. Reddit threads where the top answer is some random person guessing. The AI doesn’t sort the good from the bad. It takes it all in the same and leans on all of it when it builds your reply. So you’re not pulling from an expert. You’re pulling from the average of the internet, and the internet is mostly average.
So here’s my question. Why are we treating glorified autocomplete like an oracle?
Why we fall for it
A couple things about human psychology you need to understand.
Humans will always take the least painful path to a given goal.
Imagine someone trying to lose weight. They can put in effort and sacrifice by going to the gym and eating healthy, or they could take a weight loss pill or a GLP-1 shot.
The average human will take the GLP-1 shot or the weight loss pill.
Brian Tracy offers quite a few nuggets on human psychology in this fantastic interview:
Lets consider another example:
Picture someone who wants to build a website. One path is picking up a JavaScript book, putting in the hours, and actually learning web development. The other is typing a prompt and letting AI spit out the thing in seconds. For most people, especially anyone not already versed in it, that’s not even a decision. They take the AI. It’s just how we are.
We, the frogs, are being boiled, as they say. AI is all-consuming and its grabbing more and more every day. Critical thinking skills are quickly fading away.
It didn’t show up all at once. It crept in. First it was writing your emails. Then your code. Then your hiring, your marketing, your customer service. One task at a time, quiet enough that none of it ever felt like a line being crossed. That’s the heat going up. It’s all-consuming, slowly working its way into everything we do, and if we’re not paying attention we’ll lift our heads in a few years and find the water boiling and humans made obsolete in it.
Before AI, people studied their craft. They sharpened their blade in their specialty. They learned the intricate details through trial and error and study.
After AI, humans ask an LLM to spit out an answer based on the data it trained on. No reps. No study. Just the output.
The state of AI
Look at where we are.
We have AI creating social media posts. We have AI reading and automating interactions with those same posts. We have AI handling blog posts that aren’t even being read by humans. It’s AI talking to AI, and nobody’s in the room.
Hiring is mostly AI now too, with little human interaction. People use AI to craft personalized resumes for a given role and its responsibilities. Then HR and hiring teams use AI to filter and flag candidates. A lot of the interviewing is already AI-assisted, and it’s not hard to see it becoming AI’s job entirely. So a machine writes the resume, a machine screens it, and soon a machine runs the interview too. Where’s the human in that?
We’re using AI for everything with no clear reason. It hasn’t been thought through. Shiny-object syndrome has execs rushing to AI solutions without pairing the tool to a KPI or a goal.
AI can execute. But we forget the goal, especially in business, is to win. It’s not what you do, it’s how you do it.
Why the output falls short
Talk to anyone who’s actually an expert and they’ll tell you the same thing. AI responses are super generic. Not directly helpful in most cases.
Two reasons.
The training data isn’t great on all the variables to your given situation. And the prompt doesn’t usually provide enough context to the actual problem.
Generic in, generic out.
Follow the money
So the output isn’t even that good. Here’s the thing. It doesn’t have to be.
We live in a capitalist system. The whole game is maximizing capital and cutting expenses. And humans are expensive. Think about what one person actually costs. Salary. Benefits. Time off. Sick pay. Vacation. 401k contributions. The training to get them up to speed. I could keep going.
Now compare that to AI. No salary. No sick days. No vacation. No benefits to pay into. It runs all day and never asks for a raise.
If the goal of this society is to win the game of capitalism, humans have always been a necessary stepping stone, the thing you needed to create value for consumers. But that need is chipping away. AI is replacing the expensive part of the equation, and it’s helping the people at the top win bigger than they ever have.
They get more. You get less. That’s the trade being made, and almost nobody is being asked if they’re okay with it.
Where AI is heading
If you follow the “leaders in the space,” you’ll notice the evolution. We’ve moved from prompting to creating loops. More reliance on the AI output, less of us in it at all.
Less critical thinking. Less strategy.
Everything becomes generalized slop based on AI instructions or AI output.
Here’s your monthly reminder that you shouldn’t be prompting coding agents anymore.
— Peter Steinberger 🦞 (@steipete) June 7, 2026
You should be designing loops that prompt your agents.
Play it forward. A generation that can’t work through a problem without asking a machine first. People who never built the skill because the shortcut was always sitting right there. Whole industries running on output nobody actually understands, making decisions nobody can explain.
But here’s the part that should get your attention. When almost nobody can think for themselves anymore, the few who still can become rare. And rare is valuable. The people who kept sharpening their craft while everyone else outsourced theirs are going to run circles around the rest.
That’s the opportunity hiding inside all of this. The bar is dropping. You don’t have to drop with it.
What you should do
Pick up the book. Learn the new skill. Put that brain to work and keep it working.
Become the expert in your craft. Not the person who knows how to ask a machine, the person who actually knows. Learn how to do good work and get results, the hard way, because the hard way is the only way it sticks.
But knowing the craft is only half of it. Learn how to approach the problem and solve it toward an actual goal. Say I’m running organic content for a company. The goal isn’t page views. The goal is leads and revenue. Views that never turn into customers are a vanity number. Knowing that difference is the big picture, and the big picture is the part AI can’t see for you.
That’s how you use AI as a tool and not a crutch. You set the strategy. You own the goal. AI is the minions you point at the execution. I use this stuff every day, that’s not the point. The point is that the strategy stays yours.
Put in the work. Sacrifice for your dreams. Be the expert everyone else is busy outsourcing their way around.