I'm Done Telling People to Leave. I'm Done Telling Them to Stay.

A few days ago, my friend Annie in Kobe told me she'd been to 14 farewell parties in two months. I didn't know what to say. She said the families with better conditions have been leaving one after another this past year. To Europe, to the US, to New Zealand. Fewer and fewer are staying.

My wife just came back from a farewell party for a friend nearby. This friend had been here for years, never really had much business, and finally couldn't hold on. Going back to China. Boss Liu was more direct. He told me friend after friend running homestays are getting their visas rejected. The ones who haven't been rejected are anxious all day long. He said something I haven't forgotten. Even old overseas Chinese who've been in Japan for over a decade are feeling uneasy now.

Hearing this kind of thing too often does affect your mood. But I thought about it later, and what surprised me most wasn't that so many friends were leaving. It was my own reaction.

Our family hasn't thought about moving.

Not once.

It's like in 2023 and 2024, I kept getting asked the same question: "Why Japan?" These past six months, the question has shifted to: "Are you still staying?"

The real difficulty behind this question was never the answer. It was the motivation. The person asking is usually hesitating themselves. They want a reference from me, so their hesitation feels less heavy. If I answer seriously, they have to think about it. If I joke, they think I'm brushing them off.

At first, I used to tell some friends to hang in there a bit longer. Now, aside from telling them I'm definitely staying, I don't give any more advice.

In my first two years here, I was always saying good things about Japan on Xiaohongshu, and I used to tell friends with the right conditions, "If you can leave, leave." Later, I started telling friends who were hot-headed about coming to Japan to think twice. Only then did I slowly realize that telling people to stay and telling people to think twice are actually the same thing. Both are using my own choice as a reference for someone else.

When Annie from Kobe told me this, her tone was already very even. She's not sad. She's just seen too much. She said the people leaving include the happy ones, the lost ones, the ones who thought it through, and the forced ones. She didn't leave. She's probably one of the few around her who insisted on staying and started building an overseas business properly. I really like her attitude. Send off who needs sending off, keep who wants to keep.

A few years earlier, I used to be cynical too. Seeing the chaos at home and abroad, it would bother me. But after the frustration passed, I'd look back and realize those things didn't have much to do with me. I'm just an ordinary middle-aged man, bringing my wife and kids through life in a country that isn't perfect, taking it one day at a time. The things I can change, I can count on one hand. The things I can't change, no amount of anxiety will get me in.

So these past two years I've slowly learned one thing: don't intervene in other people's outcomes.

Take care of yourself, take care of your family. This sounds incredibly ordinary, incredibly light. But actually getting these two things right is harder than anything else.

Then there's the other side.

March 3, 2023, I landed in Japan. March 14, OpenAI released GPT-4. I was 37 that year, in the most anxious period of my work. My previous industry was changing fast, and the old playbook suddenly stopped working. I'd just brought my family to a completely foreign country, and I hadn't even started learning Japanese yet.

Every day during that period, I woke up thinking the same thing: how am I going to support this family from now on.

GPT-4 was what let me breathe during that time. Honestly, it wasn't a magic bullet, but it made me realize something. All the industry experience I'd accumulated, my understanding of business logic, my judgment of user needs, none of it had become obsolete. What had failed was how I used tools. What I needed was an upgrade, not a replacement.

I subscribed to ChatGPT as soon as I could, even though I hadn't gotten a Japanese bank card yet.

After that, Claude, Gemini, Grok came out one after another. Agent, Codex, CLI. This whole toolkit slowly took shape. Only then did I realize that my timing arriving in Japan happened to coincide with a milestone in the AI era.

I'm very grateful I'm living in Japan right now. Japan has no special restrictions on AI, and my workflow isn't disrupted by external interference. This was critical at the time. I later talked to some friends doing AI work in Europe, the US, and China. The compliance issues, regulatory discussions, and data privacy concerns they face really eat up a lot of energy. I don't have that burden. I can focus on using various AI tools.

Looking back now, those two things colliding in March 2023. You can't help but feel a bit of fate about it. I came to Japan to give myself and my wife a city we liked, to give the kids a more relaxed environment to grow up in. I ran into GPT-4 to find a way forward through a middle-age crisis. Separately, both were fine. Together, they gave my whole life a very stable fulcrum.

I'm not saying this was destiny. I'm saying some opportunities you seized because they happened to appear exactly when you needed them. This kind of coincidence looks like arrangement in hindsight, but looks like luck from the front. You can't replicate luck, but you can replicate attitude. My attitude back then was to open up and try everything, not to be picky.

I haven't changed that attitude since.

Now let me talk about something less romantic.

Three-plus years in Japan, my business is far from stable. I understood from the start that the biggest challenge for middle-aged people moving to Japan isn't the visa, isn't the language. It's whether your career curve can be redrawn in a completely new commercial society. Japan has its own rhythm, its own rules, its own social networks.

But in these three years, I haven't taken too many wrong turns. Not because I'm smart, but because I've had a few basics in place. What to do and what not to do, I have a sense of that. When things get hard, I don't waver as easily as I did in my twenties. I built myself a workflow, combining AI with specific business needs, and slowly got it running steadily.

These few things are more important than how much money I've made.

Over these years, I've also increasingly felt that the accumulation of judgment and methodology is more valuable than assets. Once you get these two things working again in a new environment, the value is reusable. Change to a different country, a different industry. When you hit difficulties, what you rely on isn't talent. It's these two things.

I also believe one thing. Chinese people, landing anywhere in the world, won't lose to local peers. We might not know the local language, might not know the local rules, but we know how to start something from zero. My wife does this better than I do. She didn't know Japanese or programming before coming to Japan. But now she can use Claude Code to automate many of the repetitive and tedious parts of e-commerce operations. From scraping, converting, generating, confirming, to listing. She can run the whole pipeline herself.

I often tell friends around me about this. A housewife who can't write code built a complete listing system just by chatting her requirements clearly with AI. It's not that AI is that amazing. It's that the Chinese spirit of figuring things out found a new outlet in the AI era.

These past three years, I've also figured something out.

Migration, at its core, is a family matter. It's not individual heroism.

When I was young, I used to think immigration was a heroic act to change your destiny. Later I realized the biggest meaning of migration is giving your family an environment where you can slowly build a good life. This environment doesn't have to be perfect, but it has to be suitable. Can your kids grow up relaxed here? Can your family live steadily here? Can the real you live at your own pace here? This is more reliable than believing any grand narrative anyone tells you.

This suitability isn't something you can figure out with a few comparisons. It requires you to actually settle down somewhere, actually live there for years, actually get ground down by the rules, the rhythm of life, the social customs. Only then do you know if it's yours.

And once you know, no matter how the policies change, no matter how the environment changes, you're not panicked inside.

If someone asks me if I regret coming to Japan, I'll still say firmly:

No, I don't regret it. But I also won't pretend this has been easy.

If someone asks me now if I'd tell others to come, I'd say:

No. You need to figure it out yourself. If you can't figure it out, don't move.

I probably won't waver anymore. And probably won't in the future either.

I only slowly came to understand something. In middle age, figuring out one thing clearly is already hard. Holding onto what you've figured out is even harder. Holding on isn't stubbornness. It's clarity. Knowing why you came, knowing what you're doing. I'm still practicing both. What I need to do next, I can't say for sure. Keep up with the changes of the times, adapt to the era, adapt to the world.

Year four in Japan, I haven't become better, and I haven't become worse. I've just become a bit more clear.

I'm done telling people to leave. I'm done telling them to persist.

I'll take care of myself, take care of my family. Getting that right is already enough.

As for whether Japan will tighten visas further, whether the world will keep shaking, whether AI will bring new uncertainty. Those are things I can't control. I can only manage my own patch.

Manage that well. The rest, I'll leave to time.