AI investors flock to Wudaokou
Wang Huiwen, the co-founder of Meituan, mentioned a magical golden circle when reviewing his portfolio.
He said, "Today, I took stock of my portfolio and made an important discovery. If you draw a square with Tsinghua University to the north, Peking University to the west, Xueyuan Road to the east, and Dazhongsi to the south, up to now, the performance of the projects within this square far exceeds those outside it. Moreover, Deepseek, Zhipu, Lovart, and Emochi, which I didn't invest in, are also within this square. I plan to suggest that my invested companies move to the center of the universe."
After that, he added, "Actually, it was the same in the era of mobile Internet. ByteDance, Xiaomi, Meituan, Kuaishou, and Didi were all within this square in their early days. The only two mobile Internet startups that made big money outside this square were Pinduoduo and Xiaohongshu. It's a pity that I learned this fact so late, which seriously affected my investment performance!"
It's indeed late because his "new discovery" has almost become an open secret in the Internet, the first - generation AI entrepreneurship, and the current AI wave. Remember years ago, China Venture Capital News wrote an article titled "The Rise of Wudaokou: The Unrestrained Growth of AI Upstarts". At that time, the AI upstarts were SenseTime and Fourth Paradigm. Although many years have passed, this title still applies, except that the protagonists have been replaced by Dark Side of the Moon, Zhipu, and Deepseek.
Who Helped "Lao Wang" Make Money
Looking closely at Wang Huiwen's portfolio, according to the official website of his personal fund Lollapalooza Capital, Wang Huiwen has invested in 23 projects so far. Among them, there are classic Internet projects such as Li Auto, Yuanfudao, and Shuidichou, as well as his well - known works in the AI field: Dark Side of the Moon, Manus, Aishi Technology, and Silicon - based Flow.
A small detail is that, different from two or three years ago, most of the projects he has recently invested in are concentrated in the field of AI applications, especially AIGC and Agent.
Among the 23 invested projects, I found that four companies, Dark Side of the Moon, Sand ai, Silicon - based Flow, and VITURE, are completely within this square. Aishi Technology and Manus are approximately within it. In total, there are only six projects.
Among these six projects, the one closest to exit and the most profitable is probably Manus.
Wang Huiwen invested in Manus in November 2024. This was the second - round financing since Manus was founded. Other participants in the same round included Sequoia China, Tencent, and ZhenFund. The post - investment valuation was 85 million US dollars. Until Manus was sold, Wang Huiwen was the only external individual investor. Just one year later, the acquisition price of Manus was raised to 2 billion US dollars, and Wang Huiwen's book return reached 23.5 times.
Although the final transaction was called off, who can say that this wasn't a brilliant move?
Excluding Manus, the project that made money the fastest is probably Dark Side of the Moon.
At the beginning of May, it was reported that Dark Side of the Moon was close to completing a 2 - billion - dollar financing led by Meituan, and the post - investment valuation exceeded 20 billion US dollars. In fact, since the beginning of this year, Dark Side of the Moon has completed three rounds of financing in quick succession, with amounts of 500 million US dollars, 700 million US dollars, and 700 million US dollars respectively. In less than half a year, Dark Side of the Moon has attracted more than 3.9 billion US dollars, approximately 28 billion RMB.
More than ten days later, an insider revealed that in an email sent to shareholders, Dark Side of the Moon did the most "counter - intuitive" thing in its two - year history: it dismantled its red - chip structure. The email said that the company would remove the offshore affiliated entities and the VIE structure, and in the future, it would submit an application for listing on the Hong Kong Stock Exchange as a domestic joint - stock company.
Previously, in the article "There Is No Older Stock Harder to Obtain Than Jieyue", we mentioned that Jieyue's process of going public in Hong Kong was significantly faster than that of Dark Side of the Moon. Now it seems that Dark Side of the Moon is also getting anxious.
Wang Huiwen first appeared on the shareholder list of Dark Side of the Moon as an individual investor in the Series C financing in 2025. This round was led by IDG, and old shareholders such as Alibaba, Tencent, and Wang Huiwen over - subscribed. The post - investment valuation was 4.3 billion US dollars, and Wang Huiwen had cumulatively invested 70 million US dollars in Dark Side of the Moon.
Although there was no news of Wang Huiwen's investment in Dark Side of the Moon in the public market before, since the angel round, pre - A round, A round, and B round of Dark Side of the Moon had investments from Chinese - American joint - venture investors or Meituan Longzhu, the outside world speculated that Wang Huiwen might have invested in Dark Side of the Moon around 2024.
However, even based on the valuation of 4.3 billion US dollars at the end of last year, in less than half a year, Wang Huiwen's book return on this project has increased by 4.6 times.
In contrast, Sand ai and Silicon - based Flow are more like Wang Huiwen's own projects. Why do I say so?
In 2023, Wang Huiwen announced that he wanted to create China's OpenAI and issued a call to recruit talents to establish Guangnian Zhiwai. He recruited two co - founders, Yuan Jinhui and Cao Yue. Later, Wang Huiwen withdrew due to illness, and Guangnian Zhiwai was acquired by Meituan. Yuan Jinhui started an AI infrastructure company called Silicon - based Flow, and Cao Yue founded Sand.ai, focusing on large video models. Now both companies have grown into leaders in their respective niches.
Finally, regarding Aishi Technology and VITURE, first, there is no information about Wang Huiwen's investment in the public data, and second, there is no latest valuation, so the specific returns are unknown.
These holdings once again confirm that even for Wang Huiwen, only a few projects make good profits. Even standing on the land with the highest density, hitting a "blockbuster" is still a very low - probability event. Among the 23 projects, only three or five can bring excess returns. Although this is the norm in venture capital.
It Still Has to Be the Center of the Universe
Let's go back to the square drawn by Wang Huiwen.
The specific scope is bounded by Tsinghua University to the north, Peking University to the west, Xueyuan Road to the east, and the north side of Dazhongsi to the south. This is a rectangular plot of about 2.5 square kilometers. In terms of area, it is less than six - thousandths of Haidian District, but it houses more than a dozen universities, more than a dozen Chinese Academy of Sciences research institutes, multiple venture capital institutions, about a dozen Class A office buildings, and dozens of technology companies.
Different from most artificially - created industrial clusters, this is not a planned science park. Instead, it is an innovation ecosystem that has grown over the past seventy years since the adjustment of university departments in 1952 and the location selection of the Chinese Academy of Sciences. It is also the center of the universe and the king of density in China's technology Internet, hard technology, and AI wave.
China is not short of science parks.
However, if we compare the indicators of this square with other technology - concentrated areas, the differences are very obvious. For example, Yizhuang has a manufacturing foundation but is far from university clusters, so it has insufficient attraction for young talents. Shanghai Zhangjiang has policy support, but the density of scientific research institutions is far lower than here. Even Shenzhen Nanshan, which has the most vibrant industries, lacks the support of top - tier university clusters in comparison.
I randomly asked several investors about this golden square, and their answer was that the resources here are excellent, and the "resources" here mainly refer to people.
Needless to say about Tsinghua University and Peking University. At the same time, Beihang University and the University of Science and Technology Beijing are also located here. More importantly, there are several core institutes of the Chinese Academy of Sciences, such as the Institute of Computing Technology of the Chinese Academy of Sciences, which is known as the birthplace of Chinese computers, the Institute of Automation of the Chinese Academy of Sciences, which plays an important role in AI vision and multi - modality, as well as the Institute of Software of the Chinese Academy of Sciences and the Academy of Mathematics and Systems Science of the Chinese Academy of Sciences.
In addition to top - tier universities and scientific research institutes, the office buildings and residential communities here also have various special meanings, and feng shui is also an added bonus. There is an old saying in Beijing that "the north of the city has good feng shui". The north of Beijing has a magnificent view and has always been regarded as a geomantically - auspicious place.
For example, Huaqing Jiayuan, which is regarded as a lucky place for China's Internet, has produced Wang Xing of Meituan, Zhang Yiming of ByteDance, and Su Hua of Kuaishou. For example, Dongsheng Building, which is adjacent to Tsinghua University, has accompanied the growth of the self - driving unicorn Momenta and the AI upstart Zhipu. Another example is Rongke Information Center, the white moonlight of Zhongguancun office buildings. It not only houses the core team of Deepseek but also witnessed the glory of Google China and Intel.
On this land, starting from Tsinghua Science Park in Wudaokou, Haidian, within a few hundred meters, you can successively see star companies and institutions such as Zhipu, Mianbi Intelligence, Shengshu Technology, and the Beijing Academy of Artificial Intelligence just by walking. If you go a little further, you can also encounter companies like Dark Side of the Moon and Deep Potential Technology.
If you are a little more careful, you can find that there is a clear pattern in the spatial distribution within this square. Universities surround the periphery, forming a wall, and the institutes of the Chinese Academy of Sciences occupy the central area, forming a core. VC firms, office buildings, and companies fill the gaps between them.
Industry and capital are latecomers. They flow into the gaps between academic institutions like water. In order to be closer to these professors, doctors, and genius entrepreneurs, in recent years, more and more VC firms and local investment promotion offices have set up their offices in the extended areas of the buildings. The clustering of these institutions is not accidental but to shorten the walking radius to Tsinghua University, the Chinese Academy of Sciences, and the laboratories on Xueyuan Road.
Thus, on this land of only 2.5 square kilometers, talents, convertible technological achievements, funds, products and services, and physical space are all gathered.
This can't help but make me think of a physical concept, "critical mass". It may not be very accurate, but the general idea is that when the density of nuclear fuel exceeds a certain threshold, a chain reaction will start spontaneously and continue.
Obviously, something similar is happening in this golden circle. A research paper turns into a patent, the patent attracts a round of financing, the financing assembles a team, and the team develops a product. This process can be completed within a one - kilometer radius and can continue to operate without external intervention.
Even during the period of overall contraction in the Internet industry from 2020 to 2023, the critical density within this square did not significantly decrease. The research institutes and universities provide a stable foundation, and they will not move due to market cycles.
So this golden square is not only a geographical square but also a high - density ecological unit of China's hard technology. Talents are here, technology is here, transformation is here, capital is here, the inheritance of generations of entrepreneurs is here, and the spirit of innovation and entrepreneurship is also here.
Wang Huiwen's square can be said to be the "geographical determinism" version of investment and entrepreneurship. Just like river civilizations can always produce food and countries earlier, the location itself determines the survival probability, and besides, the people along this "river" are indeed very smart.
This article is from the WeChat official account "China Venture Capital News", author: Zhang Xue, published by 36Kr with authorization.