This year, if VCs want to invest in unicorns, they also have to find "backdoor" ways.
A private meeting with fewer than 50 people took place in Silicon Valley across the ocean.
The theme of this private meeting was a previous round of financing by OpenAI. This round of financing, worth up to $40 billion, not only dwarfed all IPOs this year but also exceeded the largest IPO in history by more than $10 billion.
What's more worth mentioning is that not all interested investment institutions could participate in this top - tier financing feast.
The participants in this meeting were all "named" private equity investors. They included SoftBank, Blackstone, Coatue, and Sam Altman, the CEO of OpenAI himself.
The US media described this case as an "inner - circle circulation among the rich and elite."
OpenAI's most shocking financing event globally this year was a complete "closed - door feast." Only the "named" players were eligible to enter this valuation carnival.
Turning our attention to the Chinese market, not everyone can participate in the financing game of top unicorns. Especially in the later - stage financing of unicorn projects, the financial strength weeds out a large number of small and medium - sized institutions. This kind of financing tests not only the investment ability but also the connections and resources.
Cash - rich unicorns need more empowerment.
"This is more obvious in projects such as military and aerospace," said Zhang Tao, a VC practitioner. "If you can't provide some order - based empowerment, why should they take your money?"
Looking back at the history of Chinese venture capital over the past three decades, from the era when "gold was everywhere" to the current situation of "searching for a needle in a haystack" to find projects, top - tier project resources have always been scarce.
"If you have no connections or resources, don't even think about getting top - tier projects." Since the emergence of DeepSeek, investors have felt this more acutely. When a unicorn project appears, you really need some "back - door connections" to seize it.
Rare Unicorns
The sudden emergence of DeepSeek has, for the first time, so intuitively put the difficulty of investing in unicorn projects on the table.
After its brief and spectacular rise to fame, venture capital institutions were all "looking for Liang Wenfeng."
Liu Qin, the founding partner of GSR Ventures, said bluntly that more than a year ago, he sent three groups of people to try to meet Liang Wenfeng but was rejected. The reason given by the other side was roughly that they lacked innovative spirit and had different goals.
Only Chen Datong, the managing partner of Harmony Capital, was a bit special. He joked at a forum that because his investment field involved chips and had relevant business intersections with DeepSeek, he was able to communicate with its team, thus getting a little chance to make contact. Even Baidu Ventures, which shares the same building as DeepSeek's Beijing office, failed to invest successfully. Its CEO, Gao Xue, later clarified that it was because Magic Square Quantitative's large - model business was not spun off for independent financing, so they missed the investment opportunity just like other VCs.
After the Spring Festival holiday in 2025, the first research session of venture capital institutions was all about this popular company.
On February 11th, at the first weekly all - staff meeting after the start of work in the new year, an early - stage investment technology group had an in - depth discussion about DeepSeek. Why was DeepSeek so popular? What was the real technological innovation? Why was it DeepSeek?
"Who knows Liang Wenfeng? Try to find a connection and meet him." Liang Wenfeng, who was at the height of his popularity, became the "lighthouse figure" pursued by investment institutions. "If it really doesn't work, add him on WeChat. We must get in touch."
In recent years, the concentration of investment tracks and the gradual maturity of the industry have led to a visible decrease in the number of top - tier projects.
Thirty years ago, it could be described as "gold was everywhere" when VCs were looking for projects.
Xiao Bing of Fortune Capital once recalled that more than twenty years ago, he drove a Jetta and went door - to - door in Nanshan Science and Technology Park to look at projects.
Even when they found a project, many founders didn't even know what VCs did. This scene was enough to illustrate the investment environment at that time - an emerging wave, nurturing more big "fish" that were growing.
Now, the early - stage venture capital market has changed dramatically. VCs are no longer the "high - and - mighty" elite. Instead, they go to third - tier cities to look for opportunities. And the grand narrative of the past investment industry has turned into a balancing act between unicorns and LPs.
The Temptation of Sky - high Returns
The sky - high returns brought by unicorn projects are astonishing.
Take Tencent as an example. The long - term upward trend of Tencent's stock price after its listing brought continuous and considerable returns to early investors. In 1999, IDG Capital and Pacific Century CyberWorks each invested $2.2 million in Tencent and obtained 20% of Tencent's shares. In 2004, Tencent was listed on the Hong Kong Stock Exchange at an issue price of HK$3.7 per share.
Subsequently, as Tencent's business expanded and its stock price continued to rise, even with share transfers and other operations, both IDG Capital and Pacific Century CyberWorks got hundreds of times of returns when they exited. For example, IDG Capital gradually reduced its stake in Tencent and accumulated profits of more than billions of dollars. This investment also became a classic case of its investment in the Internet field.
Looking more recently, the listing of Moore Threads, the leading domestic GPU company, has attracted attention to early - stage investments in the primary market. This is a star unicorn project on the Science and Technology Innovation Board. Its valuation was only 480 million yuan after the angel round in 2020, and its market value reached 332 billion yuan after it was listed on the Science and Technology Innovation Board in December 2025, with a valuation increase of more than 690 times. If Heli Capital and TEDA Science and Technology Investment, which participated in the early - stage investment, had held their angel - round investments until now, their book returns would have exceeded 690 times, and currently, their actual book returns have already exceeded 100 times.
After Ge Weidong, a private equity tycoon, entered the game in February 2025, the book value of his personal shares was about 11.9 billion yuan, with a floating profit of more than 11 billion yuan, nearly 14 times the investment. The total investment of his Chaos Investment was about 649.2 million yuan, and the book value of the shares exceeded 10.4 billion yuan, 16 times the investment cost. In addition, early - stage institutions such as Sequoia Capital, ZhenFund, and Matrix Partners also reaped rich rewards.
Moore Threads, known as the "Chinese NVIDIA," is a unicorn in the domestic GPU field. It completed two rounds of financing only three months after its establishment in 2020, and its valuation exceeded $1 billion in less than 100 days. The initial investment of only 1.9 million yuan by Peixian Qianyao, an early - stage investor, had increased to about 11.898 billion yuan when the company was listed on the Science and Technology Innovation Board in 2025 with a market value of 280 billion yuan, a return of more than 6,262 times.
Subsequent institutions such as Shenzhen Capital Group and Sequoia China, which participated in the Pre - A round, and Shanghai Guosheng Capital, which participated in the A round, also obtained substantial returns due to the continuous increase in the company's valuation.
One phenomenon is that it is becoming increasingly difficult to get a seat at the table of top - tier unicorns.
Good - faith Deposits and Role Reversals
Two or three years ago, there was a new thing quietly circulating in the venture capital circle - it was not the investors who selected projects, but the project parties who set "exams" for investment institutions.
"They just sent over a questionnaire. We not only had to introduce the details of our own institution but also answer essay questions about our understanding and views on the project and the industry," said Zhang Tong, spreading his hands with a tone full of helplessness. "Only after submitting the questionnaire could we be eligible to talk about the next step."
What was even more frustrating was not even getting the chance to submit the questionnaire. Zhang Danyang sighed: "We finally found the contact information of the project party and applied to add them as a friend, but there was no response at all, not even a prompt indicating that the application was approved."
In this game in the venture capital circle, not being able to find the contact information of the project party was just the first setback for investors. An even more helpless second threshold was being directly rejected for due diligence.
Previously, a leading project in the manufacturing field announced a financing plan. This company ranked first in the global industry. As soon as the news came out, it instantly became a hot target in the investment circle, and investors from all over rushed to get a share.
To establish connections, everyone used their own methods: some asked the founder's university teacher to make an introduction, some used alumni to pass on messages, some asked previous - round investors to recommend them, and some even asked government departments to act as a bridge. All kinds of personal connections were mobilized just to get a little investment quota.
"We finally managed to get through the barriers and were about to enter the due - diligence stage, but they directly rejected us," Wu Deji still felt a bit helpless when recalling this.
It turned out that this project put forward a financing plan of "equity + debt" and required all potential investors to bid. The rules were very strict: only by subscribing to a certain proportion of interest - free debt first could they be eligible to invest in Class B shares. Even so, the temptation was irresistible - more than a hundred investment institutions didn't hesitate to pay the deposit and submit the bidding application. "At that time, each institution paid more than $100,000 as a good - faith deposit."
What's more interesting is that because the project was so popular, many middlemen emerged in the market, trying to make a profit by acting as a go - between. In response, the project party simply issued a risk warning: "Don't waste money on middlemen or ask internal employees to put in a good word for you. Doing so will only deduct points from you and is just a waste of money."
The project party had the confidence to be so tough because of its absolute advantage in the market. Wu Deji also encountered another similar project - a well - known medical unicorn. "This one was even more extreme. It directly announced that it would not open for due diligence. Whether you were a first - tier top - level institution or a second - or third - rate small institution, it treated everyone equally and didn't show its cards to anyone."
Wu Deji said with a bitter smile when looking back: "Not allowing due diligence really stepped on our investment bottom line. In the end, we didn't invest in either of these two projects."
The ending of the story didn't see the reversal that investors were expecting. Now, these two projects that once kept people at a distance have both developed quite well. One of them has even established a firm foothold in the global market and built a strong industry influence.
For a long time, the voice in the venture capital circle seemed to be in the hands of investors - they selected tracks, evaluated teams, examined financial statements, scrutinized each project with the magnifying glass of due diligence, and held the initiative in negotiations with valuations and terms. But now, this phenomenon has been completely broken.
A well - known figure in the industry said about a newly - listed project worth hundreds of billions that he invested in: "I know that he (the founder) deals with a lot of WeChat messages every day and usually starts replying at two or three o'clock in the morning. So I send him messages at eleven or twelve o'clock at night to make sure my messages are at the top."
When a project has absolute barriers in technology and market share, it has the confidence to conduct reverse screening. It doesn't care about the reputation of the institution, doesn't accept exaggerated flattery, and even dares to pause the most basic due - diligence process. Behind this is the confidence brought by product strength and industry status.
Good projects are scarce, but when faced with such a project that holds the initiative, investors find that their past investment logic may need to be re - evaluated. Those due - diligence processes and risk - control standards that were once regarded as the golden rules seem a bit powerless in the face of absolute strength.
"Invitations"
Such stories have long been playing out across the ocean.
OpenAI's "closed - door financing" is no longer an isolated case. In the top - tier investment circle, the real good opportunities are never put on the public shelf. For those leading projects that can disrupt the industry and create a market value of hundreds of billions, it's never the investors who select projects, but the projects that "screen" investors.
Money? It's just the entrance ticket. The real pass is whether you are on the founder's "trust list."
The giants on Wall Street have long understood this rule. Morgan Stanley, JPMorgan Chase, and Goldman Sachs have all established exclusive private - market departments. Their service targets are only of one type: institutions, family offices, or a few recognized wealthy individuals that hold huge amounts of capital and can establish connections with top - tier entrepreneurs. These people don't need to scramble for shares in the public market but wait for the "invitations" actively sent by the project parties - just like a private party. Even if you meet the wealth requirement, if you're not on the list, you'll be blocked outside.
Investors who miss these closed - door deals have to settle for the complex and expensive private secondary market. Through a complex structure like SPV (Special Purpose Vehicle), they indirectly hold the shares of these non - listed companies. But there is no transparent pricing and no smooth liquidity here. It's more like a gamble, using a high cost to buy a vague future.
Among all the players in these "invitation - only games," SpaceX is undoubtedly the most dazzling one. Recently, there have been widespread rumors about its upcoming IPO, but before that, it had already completed a "rocket - like launch" of its valuation in private.
This summer, in a private share transaction, SpaceX's valuation had risen to about $400 billion. The participants were mainly Musk's long - term supporters, early - stage investors, and institutions "named" by the company.
Just a few months later, an internal memorandum issued by SpaceX's Chief Financial Officer, Brett Johansen, excited the entire investment circle: the company's latest internal stock - issuance price was set at $421 per share, corresponding to a valuation of up to $800 billion - this was not only twice the valuation this summer but also directly exceeded the $500 billion record set by OpenAI in October, reclaiming the title of the world's "King of Unicorns."
One of the biggest beneficiaries of this valuation leap is Google.
This technology giant, which bet on SpaceX in 2015, together with Fidelity Investments, invested $1 billion and obtained about 10% of SpaceX's shares. This decade - long patient investment has already started to bear fruit. At the end of last year, after SpaceX determined a valuation of $350 billion through a tender offer, Google quietly disclosed an "unrealized gain on non - listed equity securities" of $8 billion in its April financial report this year - the market almost immediately guessed that the source of this gain was SpaceX. And this unexpected windfall directly made Google's first - quarter net profit exceed Wall Street's expectations.
Now, as SpaceX's implied valuation soars to $800 billion, everyone is watching Google's next financial report. How many billions of dollars will the book profit be this time? No one knows the exact figure, but it's certain that Google's initial investment of $1 billion has already turned into an astronomical figure.
The memorandum also hides a greater ambition: SpaceX is intensively preparing for an IPO in 2026, planning to raise more than $30 billion. Musk's goal is even more astonishing - he hopes that the company's overall valuation can reach $1.5 trillion, approaching the world's highest market - value record set by Saudi Aramco when it went public in 2019.
If this goal is achieved, Google will receive an unimaginable sky - high return. And all of this started with an investment recognized in 2015, a permanent pass to enter the "invitation - only market." In the world of top - tier investment, the most expensive thing is never money but trust and patience.
The venture capital circle has never been a one - way - choice game. When entrepreneurs no longer need the capital from investors to "transfuse blood" for survival but can "generate blood" and lead the way with their own strength, they are qualified to re - define the rules of the game.
This may also be a sign of the industry's maturity - a truly good project never begs for capital's favor but is chased by capital. And investors also have to learn to adapt to this role reversal, find a new way to communicate with top - tier projects while sticking to their bottom line.