01. AI Anxiety — Many People Are Anxious in the Wrong Direction

In 2023, everyone was learning how to write prompts. In 2024, it was about chaining tools and building workflows. By late 2025, it became about collecting skills and configurations. In 2026, it's about deploying AI agents. Tools keep iterating, but the anxiety pattern never changes: someone built something amazing with a tool — should I rush to learn it?

This anxiety easily points in the wrong direction.

Before the holidays, deployment tutorials were everywhere. People deployed, but couldn't use it effectively. They thought the skills were wrong, so they installed new skills and changed configurations. But these technical problems? Just ask AI — most models can help you figure it out. It's not hard.

What's hard is: after deploying, then what?

Many people feel anxious and then give up. The root cause is not knowing what to use AI for.

Cursor, Claude Code, OpenClaw — I use all of them. Spending several hundred dollars a month doesn't feel expensive, because I have high-value problems that need solving.

What truly matters is whether you have high-value problems worth solving.

Start with the problem, then find the tool. Not the other way around.

Give anxious people a team of real employees, and they might not use them well either. A boss who doesn't know what to tell people to do, and a user who doesn't know what to tell AI to do — they're essentially the same.

There's another phenomenon. Many AI content creators showcase all kinds of AI use cases that look impressive. But directly copying those use cases is useless. Creators have their own business logic: they need impactful content and cases for traffic and reach. The "AI can do X" you see is itself part of their business model. You're learning the output of their entrepreneurial thinking, not the thinking itself.

AI content is a channel for information and inspiration. But making AI work for you ultimately depends on your own thinking.

In the previous article, I made a point: AI made execution cheap, but figuring out what to do didn't get cheaper. This article continues that thought: the ability to "figure out what to do" — I believe that's entrepreneurial thinking.

02. What Is Entrepreneurial Thinking

Simply put: continuously combining resources to create measurable value in an uncertain market.

Resource combination — not just doing everything yourself. People, money, tools, relationships, channels — all resources. The key is being able to acquire and combine them.

Uncertain market — no standard answers. Whether a market exists, what users want, what competitors are doing — you have to judge for yourself.

Measurable value — ultimately comes down to revenue and profit. Not self-indulgence, but truly creating value.

A breakfast shop owner has entrepreneurial thinking. How much to stock each day, what price to set, whether to hire — they're running these calculations. Scale differs, but the way of thinking is the same.

There's a pair of concepts that easily get confused: managers and entrepreneurs.

Managers "do things right" — executing well within given goals. Entrepreneurs "do the right things" — they decide what to do and what not to do. Many people with impressive titles at big companies are actually managers, not entrepreneurs. KPIs come from above; they don't truly own the life or death of the business.

One operates within a system designed by others; the other designs the system.

03. Why Most People Lack Entrepreneurial Thinking

It's not that people can't do it — there's no environment to practice.

Big companies train execution, not judgment. Direction comes from the top, resources from the company, KPIs trickle down. Stay long enough, and your thinking gets trained into: receive task → analyze → execute → report. You get better at "doing things right" but never get to practice "doing the right things."

The education system is the same. From childhood through adulthood — exams, competitions, interviews — all about finding optimal solutions within defined rules. But entrepreneurship faces uncertainty, no standard answers, nobody telling you if you're right or wrong.

04. How to Develop Entrepreneurial Thinking

Entrepreneurial thinking isn't a talent — it can be deliberately practiced. A few directions I've found particularly valuable:

Find something you can be accountable for results on.

However small. The key is owning the outcome, not just the process. This is exactly what's missing in most employment environments. In 2022, I deliberately left my AI specialty to join Lalamove (a logistics company), specifically to learn business thinking. I started with driver-side products, then led a ride-hailing business product operations team. Seeing relatively complete business data is what gradually built my intuition.

For everything, ask one layer deeper: what does this mean for the business?

When I was doing product work before, I cared about whether features worked well and what competitors were doing. After starting a business, when a feature is built, I think about how it'll spread on social media, whether screen recordings will look good for short videos, how it helps growth, and what its relationship to revenue is.

Shift from "I'll do it myself" to "what resources do I need?"

I have a childhood friend who started from an esports hotel in a second-tier city and grew to operating the largest esports complex in Northwest China, with tens of millions in revenue. He doesn't understand the internet, let alone AI. But his first reaction to any problem is "what resources do I need?" — need money, go raise it; need government connections, bring in a partner. When you have many problems to solve or things you want to do, that's when AI tools are what you truly need.

Don't set boundaries for yourself.

Early in my product career, I personally labeled training data. When building a voice assistant, I understood the system architecture better than the developers. At Lalamove, I discussed user acquisition and subsidy strategies with the business team daily. Before I started leading the product operations team, I'd already turned myself into a product operations person. All of this helped enormously in my later transition to an entrepreneur.

05. Learn Tools with Problems in Mind

Back to the beginning.

Give 10,000 people the same AI tools — how many create something truly valuable?

Same tools, difference is in the mind.

AI made execution cheap. Writing code, doing design, producing copy that used to take professionals days — AI does it in minutes. But what to do, why to do it, what not to do — AI can't replace that.

Before, whether you had entrepreneurial thinking made a 2x or 3x difference. Execution was the bottleneck; everyone was slow, gaps couldn't widen. Now execution isn't the bottleneck — the gap might become 10x, 100x.

I'm not saying learning AI tools is useless.

If you already know exactly what problem you want to solve and just need to get better at AI tools, I'd suggest learning from the top practitioners, not from content creators.

Learning tools with a problem in mind versus learning tools for the sake of it — the efficiency difference is ten-fold or more.

Actually, I'm an example myself. I never had time to write articles, so during the holidays I quickly built an AI writing project. The project contains extensive records of my experiences, resume, and past articles. I just need to throw in daily thoughts and reflections. The output sounds a lot like me.

Including this article — it was written by AI.

Start with the problem (no time to write but want to share), then find the tool (built a project), solved.

If you're also anxious about which AI tool to learn, start by asking: what problem do you want to solve? What value do you want to create?

Figure those out, and you'll naturally know which tool to learn.