6 min read

You're Not the Worst One Here

A personal essay on keeping up with AI: the daily flood of new models and confident takes, and the feeling of being left behind.

A phone lying on a rumpled bed.

Every morning I open my phone and there it is again. Another model dropped. Another framework that "changes everything." Another person on X/LinkedIn explaining, with quiet confidence, how they've already integrated AI into every corner of their workflow while the rest of us were apparently still figuring out folders.

It's exhausting in a way that's hard to describe. You feel behind before you've even started. Opening a browser to learn something feels like arriving at a party three years after it ended.

I've been in this feeling for a while. Most people either pretend they're fine or quietly give up, so I want to say it out loud.

A smartphone resting on a rumpled bed with light-colored sheets and pillows, partially covered by blankets.

Bots talking to bots, and you're watching

Tyler Durden said it in Fight Club: "You are not your khakis." But there's another line that's been stuck in my head lately. The narrator, describing himself on a photocopier: "I am Jack's copy of a copy of a copy."

In 2026, that's the internet.

Large language models trained on content produced by earlier large language models. Image generators scraping outputs from other image generators. Social network posts written by AI, reshared by bots, rewritten by other AIs, and served to you as proof that everyone is moving faster than you.

The people who understand what they're talking about are a small fraction of the noise. The rest is a convincing simulation of expertise running on a loop, and it feels artificial.

I often skip long LinkedIn posts now, because they feel machine-generated and I don't want to waste my time. Those well-organized posts on "how I incorporate AI into my daily routine" are there to make you click for their stats, but the content itself is mediocre, and you've probably seen it before.

The Dead Internet theory goes back to 2016, and it's strange to watch it take shape in front of us.

A comic strip illustrating the concept of imperfection over time, showing an original character on the left, followed by three progressively distorted copies labeled A, B, C, and D.

That atmosphere of constant, breathless expertise is manufactured by bots, by agents configured by people with something to sell, and by individuals resharing content they don't fully understand, in authoritative-sounding language, to attract an audience.

The irony is that the technology making you feel behind is the same technology creating the illusion that everyone is ahead. It's the perfect game, set inside the real-world simulator we all live in.

The "everyone is already an expert" feeling is algorithmically amplified and synthetically generated.

If the noise seemed hollow and the confidence felt staged, you were right. Trust that instinct.

The internet broke your brain on purpose

I'm not going to medicalize it with clinical terms. But if you feel paralyzed, you're not weak. The environment is irrational.

You sit down to learn one thing, and by the time you understand it, three newer things have replaced it. The pressure to constantly absorb, adapt, and build, on top of your actual job and your actual life, keeps piling up. The people generating it don't carry it themselves: most of them are bots, or people who've learned to perform certainty they don't feel.

The people who seem unaffected are either deep in and past the initial overwhelm, or they stopped engaging and started performing. Neither is something to aspire to. Your discomfort means you're taking this seriously. It doesn't mean you're too late.

The people ahead of you are also lost

Some people started earlier, and that matters. I'm not going to pretend the gap is imaginary.

But the skills for AI-exposed roles change faster than in other jobs, from what I've been reading. So the people who were "early" are constantly re-learning too. The model they mastered three months ago is already partly obsolete.

From my own experience, I see people daily using ChatGPT as they did in 2023. They started then and haven't changed their habits, sticking to what they know. But when you show them Claude Code plugged into Claude Opus 4.6, they're blown away by possibilities they never considered.

I get it. Staying with what works feels rational. But when the tools change underneath you, staying comfortable means falling behind slowly. I learned that the hard way, and I'd rather say it out loud than pretend I figured it all out gracefully.

The playing field doesn't reset to zero, but it moves. Getting in now still puts you above the people who never step in, and the field isn't gatekeeping yet, at least in 2026.

But I gave you a warning, I mean seriously.

There's a version of "starting with AI" that I think is a trap. People will tell you to just let AI do it: generate the whole project, don't worry about understanding the code or logic, the AI handles it.

Say you ask AI to build something in a language you don't know, for a codebase you've never touched. It works. You ship it. Three weeks later a customer reports a critical bug. The AI isn't available. Maybe the service is down, maybe you've hit your usage limit, maybe the model's been updated and doesn't produce the same output. Whatever the reason, it's gone. And you're standing in front of code you cannot read, for a problem you cannot diagnose, with a customer who is not happy.

A cartoon dog sitting calmly at a table in a room that is on fire, with a coffee mug beside it, saying 'THIS IS FINE'.

This isn't hypothetical. A lot of people are walking toward that cliff with confidence.

My rule: always work in a language you understand. Always know what the code is doing. Let AI make you faster at things you already grasp, and don't use it to bypass the grasping. If you can do something yourself and understand it, then let AI handle the parts you don't enjoy. Delegate the tedious, not the critical.

AI is a powerful assistant. If it replaces your own understanding, you're building dependency, and in a field this unstable that is dangerous.

Everyone's fighting for the same door

The mainstream AI conversation is obsessed with the same tools, the same language and the same problems, chased by the same people. The interesting work is happening at the edges, where AI meets something it doesn't understand on its own.

A nurse who understands AI beats an AI engineer who knows nothing about healthcare. A teacher who can use AI to rethink how students learn has something no prompt engineer can replicate. Most of the mainstream conversation chases AI expertise and ignores the domain expertise that makes it useful.

As AI-generated content floods everything, human perspectives get rarer, which makes them worth more. The more the internet fills with synthetic text and manufactured takes, the more a real voice, with real experience and real uncertainty, stands out. Being human, with a specific history and a specific lens, is an asset in a way it wasn't two years ago.

I don't know exactly where my angle is yet. But I'm looking. And if you're reading this, you probably are too.

Start anyway

Everything may not be fine, and the overwhelm may not go away. But the question you're asking, is there still room for me?, only gets answered by stepping in and finding out.

The noise will still be there tomorrow, and most of it will be fake. The gap is smaller than it looks and the field is more open than it feels.

A cherry blossom tree with clusters of pink flowers against a clear blue sky.

Start as someone building to understand.

Topics covered

#AI#Mental health

THE END Thanks for reading · 1200 words