I don't know. I'm using AI.

English | Chinese Original

English translation of the Chinese original. This version is generated for international readers and may be refined over time.

Dates: 2026-03-14

Just yesterday, I sent out an article that looked at the ssh login, and then some readers and friends left messages and some tweets consulted me about my technical details, which gave me some insight.

I've learned nothing about technology, I'm using AI, and I can't say it all.

Now everyone knows a phenomenon that AI brought technology parity. Whoever it is, which industry, whatever level, if you can optimize Prompt, get the demand right, and when it's over, AI will give you what you want. The effect is not silky, it's not what you want, it's entirely dependent on your hints and LLM model capabilities.

But yesterday, when ssh was checked and somehow killed, something different was found.

AI can only die within one route, which means that his context determines that he can only go one way to the dark, and he will not jump out of this framework to look at this in the larger picture.

The shortest line of distance between the two points may be a simple one, but he's got you around and doesn't understand. That means you need to show him the way.

I saw a friend who left me a message saying that if he were to re-assemble the system in such a situation, it was a little technical novel.

But I happen to know a little bit about technology, but that's it, and it's used in this sort of process.

In other words, AI is the top solver, but he's not the picker. AI will magnify your technical skills, but it will not replace your decision-making.

Friends who don't understand technology may use AI to solve 99% of the problem, but it's possible that this 99% is wrong from the point of view. So the last 1% may not exist.

I've heard and seen a lot of articles lately, and everyone's actually worried. Anxiety was replaced by AI, unemployed, lost in the future.

In fact, there are concerns that many of the job attributes will not be replaced, but will be transferred.

And it's really good for AI, so that the problem can be solved efficiently and fundamentally, and it still depends on people's technical thinking, scrutiny experience and systematic judgement.**

And I'm sure you've heard a lot about this wave, "Fuck, this big model is awesome," "half a day, from 0-to-1 online, a product," "Ai, give me a night to finish."

Most people finish their product, throw it away, there's no iterative properties.

On the one hand, there is no idea how to optimize, and on the other, there is no idea what to optimize, which requires an iterative and optimal thinking.

Instead of being replaced by AI, think of how to hold and magnify this 1%'s irreplaceable. Technology equity is the right to use tools, and inequality is about human judgement, creativity and systemic thinking. This 1% is not the end of technology, but the starting point for us as human beings.