Maybe you should take a look for yourself first.
This is a little challenging for Westerners; the Chinese internet ecosystem is almost totally unknown to us. Re: Thoughts Memo, would this be the correct link? I must admit I've never heard of Zhihu, but it appears to be a combination of Quora[1] and Substack and AO3?
Clicking through links a little there, I found this pretty funny post[2] (not by Thoughts Memo):[3]
Yeah, that sounds like me. I also notice your post here has no links (nor even LW's famous footnotes!). Interesting cultural difference!
Anyway, I can understand why Anthropic/America's fear and distrust of the Chinese Communist Party, and China as a whole by extension, would negatively polarize Chinese people against rationalists/effective altruists.
I can say two things in our defense:
- We're really afraid of AI permanently (literally forever) entrenching governments that will strictly control what their populaces can do or say. We tend to think humans can only truly flourish without such restrictions, even if that approach does mean that ex. the United States is a chaotic, politically-charged mess of contradictions even on its best days.
- Rationalists, and especially effective altruists, have always believed in the goal of a better future for all human beings, not just Americans or Westerners. That definitely includes Chinese people! We're also more pluralistic than sometimes is appreciated; many of us want something like multiple simultaneous utopias that people can choose between themselves.[4]
Also personally, my feelings on the CCP aside, I have a pretty positive view of China and its people, not to mention a great deal of respect for them. I also have a little understanding of Chinese history and culture. But even then, attempting to speak on Chinese platforms feels intimidating, as there's clearly tons of cultural context I don't have, from the chengyu to simply what it feels like to go through the Chinese educational system. Also, I understand that there's often requirements to log in with a Chinese phone number. Maybe there's a good way to engage, but we need more instructions and directions to find it.
- ^
Link to random question to bypass the terrible Quora homepage.
- ^
Note, you can just click out of the login prompt. To read, you can use your browser's translation functionality or paste into an LLM.
- ^
Translated by Chrome's translation feature.
- ^
This, too, is also a core American value IMO - individual states in the United States were originally thought of as being much more independent and different from one another, and states were founded somewhat experimentally at times, most notably Utah, which was founded as effectively a breakaway religious commune. The original 13 colonies were fairly experimental themselves, too.
Uh, actually, I'm the one who posted that pretty funny post! I only posted that one article on Zhihu to test the waters. Most of my translations are hosted on a relatively niche web novel community (including 7 articles, 11 short pieces of rationalist fiction, and a few podcast episodes about rationalist fiction).
The reason there were no images or hyperlinks is simply that I ran out of time to polish it yesterday. I spent my entire evening after work writing it, and by the time I hit publish, I had less than 5 hours left to sleep. It was the first long-form original piece I've posted online this year. So when I say I'm not good at certain things, I really mean it.
1.Many Chinese people share similar concerns, as dealing with platform censorship is part of our daily routine.——I mean, sometimes you just need to watch your phrasing, or the platform's auto-moderation might take your post down
2.While many Chinese people understand the context, some might over-interpret it and associate it with political topics.
I totally understand your apprehension. Maybe you could check out some foreign content creators who are currently active on Chinese social media. A phone number might be an issue, but I'll look into it and see if I can help. By the way, I'm not entirely sure if a foreign IP address affects how the platform pushes content—perhaps using a VPN might be necessary?"
1
1Thoughts Memo is not well-known beyond China. Maybe FSRS (a spaced repetition algorithm) is more well-known than it. Both of them are my project. I put my effort on education technology & learning science before, until I found that AI may destroy everything. Most AI users just let AI take over their work, and they stop from learning. I'm translating AI safety articles since last week with my friends from Thoughts Memo.
2
When I finished HPMOR, I immediately knew it was the best novel I had read in more than a decade. I only wished I had found it sooner.
When I started reading The Sequences, I discovered that the Chinese translation group had translated only the first volume. When I graduated from university, two years ago, AI translation had only just become good enough to convey the meaning of an article with reasonable accuracy. It was only about a year and a half ago that I truly found my way here and began engaging seriously with rationalism.
My score on the Chinese college entrance exam was only slightly above the cutoff for what was then called a first-tier university. At university, my grades were near the bottom of my year, and I almost failed to graduate. It is probably fair to say that the vast majority of graduates from first-tier Chinese universities are smarter and more capable than I am.
English has always been my worst subject. From childhood through school, I could barely pass it.
I have now been working for two and a half years and have saved about $15,000. That is enough for me to cover roughly five years of basic living expenses, including minimal insurance but excluding rent.
If there is no major good news about AI safety in the next few months, I plan to quit my job before February 2027 and devote myself full-time to work related to AI safety. For now, I already spend most of my time outside work, as well as whatever spare time I have during work, on it.
Two months ago, I learned how to typeset comics in Photoshop and completed a preliminary Chinese typesetting of the HPMOR comic. When GPT-6 was released, I paid for an AI service for the first time—not counting AI translation—and then began working on two projects.
The first was an attempt to build a workflow that could take the preliminary typeset files of the HPMOR comic and automatically generate an editable first-draft PSD in any language, ready for human correction and proofreading. I got version 1.0 working, but found it difficult to balance reducing human labor with reducing token usage. I am currently rebuilding it as version 2.0.
The second project was based on saving AI-translated rationalist webpages with an HTML downloading extension. I built a frontend that allows Chinese readers to browse the AI-translated versions of those pages. It includes an on-demand terminology glossary, topic categories, search, and reading guides that connect articles according to different reading motivations.
The basic functions of version 1.0 are already working. It still needs UI improvements and a sufficiently large collection of imported articles.
The purpose is to let interested Chinese readers reach the information they need with as little friction as possible.
Before this, I had proofread Chinese translations of rationalist articles and posted them on forums and Zhihu. But manually dealing with webpage-specific formatting—special fonts, hyperlinks, and so on—was far too inefficient.
Later, I would also like to try making an HPMOR audiobook, AI videos, fan animations, and similar projects. I have not started those yet, and I have not asked EY for permission.
I also want to make attention-grabbing introductory videos about AI safety, as well as biographical or profile videos about EY and other people in the rationalist community.
I am not good at editing, making videos, writing articles, or programming. I know that I really am the kind of person who is neither especially smart nor especially good at getting things done. I will do everything I can to accomplish something, but I also suspect that my efforts may not achieve very much.
The reason I am doing these things is that I currently know almost nothing about the technical side of AI safety. Learning enough to contribute there would take me months, and right now I feel that we may not have that much time. I also have no particular advantage in mathematics or computer science.
Meanwhile, on Goodreads, HPMOR has roughly ten times the numbers of The Sequences. I also looked at the 2022 and 2023 LessWrong community surveys, and roughly 30% of respondents said they had found their way here through HPMOR.
I have read posts on LessWrong about China, and many people's impressions of China are very different from my own.
I think one thing is important to understand: China went through, in a few decades, changes that took some other countries centuries. Because of that, the differences between one Chinese generation and the next can be enormous.
Since Jacob Coxon quit his job, I have watched AI safety become a major topic in the West and then slowly, piece by piece, spread into every corner of the Chinese public sphere.
Every day, every minute, I have watched one-sided attacks and distorted portrayals of rationalism,[1] rationalists, and the AI safety community become more and more common.
I do not know how much time we still have to act.
We cannot let people who pander to their audiences for money and say whatever their audiences want to hear come to dominate public discourse.
At the very least, dismissing x-risk shouldn't be the only narrative in circulation.
Once a narrative hardens into consensus, the cost of reversing it becomes absurdly high.
There is a saying in China:
“If you do not occupy the high ground of public opinion, your enemy will.”
The problem is not merely that you lose help you might otherwise have received. You also create additional resistance.
There are many Chinese people who are internationalists. In fact, within much of the Chinese left, internationalism is regarded as far more correct than nationalism.
There are many Chinese people who genuinely hope for a more peaceful and better world.
There are many Chinese people who know nothing about rationalism, transhumanism, or effective altruism, yet who would still be willing to work themselves to the limit to help others—even to sacrifice themselves.
They are smarter than I am.
They are more capable than I am.
They can do things better than I can.
AI is now good enough for us to enter each other's real public spheres and see what people are actually talking about.
There will inevitably be cultural differences. But underneath them, human beings are fundamentally alike.
Even if you do not speak a single word of Chinese, with AI you can still go among them and show them a real version of yourself—just as I am here, talking to you now.
Letting people see one real, living human being can be enough to outweigh a long list of negative labels placed on them by someone else.
China produces about 12 million graduates every year, including roughly 5 million graduates from first-tier undergraduate institutions and around 3 million graduates in science and engineering fields.
China has around 350 million educated young people between the ages of 10 and 30.
They know almost nothing about rationalism, transhumanism, or AI alignment.(The Dark Enlightenment, though, is popular.)
Please give them a little hope.
A little support.
A chance.
And they may give humanity a little hope, a little support, and a chance.
---------
I have to go to sleep now.
I know this post is not very clear. But knowing myself, if I do not publish it today, I may never publish it at all. I will try to improve it tomorrow[2].
I also want to show you some polling data about HPMOR。
I also want to introduce you to Thoughts Memo, the translation group that took on the Chinese translation of The Sequences. They are now publishing translations of The Sequences on Zhihu every day, along with translations of essays by EY and other rationalists.
There is still so much more I want to say.
But I really have to sleep now.
Maybe you should take a look for yourself first.
1：07：45—1：15：50，The speaker summarizes the rationalist movement, as he understands it, in terms of several claims:
Some of these characterizations are clearly exaggerated and framed in a negative way.
Tweaked the formatting, fixed a few mistranslations, added an image and some links. On second thought, I decided to leave the bulk of the content as is.