China's "lobster farming" is causing such a sensation that the whole world is amazed. Goldman Sachs sales are astonished: the speed and enthusiasm of the Chinese!

China's "lobster farming" is causing such a sensation that the whole world is amazed. Goldman Sachs sales are astonished: the speed and enthusiasm of the Chinese!

An open-source AI agent software has sparked a nationwide "raising lobsters" craze in China, astonishing the world.

A sales memo sent out from Wall Street has put China’s "lobster-raising" spectacle in front of global investors. In the memo, Goldman Sachs Asia sales Philip Sun wrote that almost everyone in China is now "raising lobsters," and the speed and enthusiasm with which Chinese people embrace AI is truly breathtaking.

He wrote bluntly in the memo: "Chinese AI company MiniMax set my personal record—I’ve never participated in an IPO project that grew 740% in just two months." Philip Sun believes the driving force behind this is deeply connected to the "lobster-raising" frenzy sweeping across China.

Meanwhile, on the US Reddit forum, a post about a long queue at the entrance of Tencent’s Shenzhen headquarters for machine installation received over a thousand likes, and the comments were packed with bewildered foreign netizens: "This scale is insane, right?"

Goldman Sachs Salesman "Breaks Down": Never Seen Such an IPO in My Life

On March 11, 2026, Philip Sun, sales from Goldman Sachs (Asia) Global Banking & Markets, sent a memo to clients titled "Train your own 'Lobsters'". The very first sentence ignited the market:

"MiniMax set my personal record. I’ve handled so many Asian IPOs, never seen a stock that surged 740% in two months."

The numbers are straightforward—MiniMax was listed at HK$165 on January 9, 2026, and traded at HK$1,220 by March 11. Two months, 740%. At the same time, another Chinese AI company ZhiPu AI was listed on January 8, with a two-month increase of 560%.

(MiniMax weekly chart source: Wind)

Philip Sun also mentioned in his memo that their management roadshow began on December 29, and even the last roadshow in Hong Kong at 3 pm on December 31, New Year's Eve, was full—finance people were still meeting and fighting for projects during the new year.

This veteran, who has seen countless Asian IPOs, couldn’t hold back several exclamation marks in his memo, using the term "Quite amazing".

MiniMax is a company with an average employee age of 29, and its market value now equals Baidu's.

Nearly a thousand people lined up outside Shenzhen just to install a "lobster"

What impressed Philip Sun wasn’t just the share price. The other scene he described in the memo is what truly triggered global attention—

Tencent employees set up stalls outside the headquarters building in Shenzhen, offering free help to install OpenClaw on people’s computers.

What is OpenClaw? It’s an open-source AI agent software developed by Austrian programmer Peter Steinberg, and its logo is a lobster.

Unlike ordinary chatbots, OpenClaw is designed to be truly "capable" AI—it can clean emails, handle files, write code, post to social media, and even complete complex tasks automatically 24/7.

Even Nvidia CEO Jensen Huang said at the Morgan Stanley Tech Conference in San Francisco: "OpenClaw might be the single most important software release in history."

This "lobster" entered China and immediately exploded in popularity.

At that event outside Tencent headquarters in Shenzhen, nearly a thousand attendees included retired aerospace engineers, housewives, students, and all sorts of AI enthusiasts—a scenario with so many "atypical users" simultaneously present speaks volumes.

Shanghai designer Mark Yang said after using OpenClaw, it felt like he had a batch of "virtual employees", and many tasks no longer needed his involvement.

Entrepreneur Fu Sheng, while bedridden during Spring Festival due to a skiing injury, used 14 days to build an AI assistant called "Sanwan" based on OpenClaw:

In 4 minutes, sent New Year’s greetings to over 600 friends, posted social media content that reached a million reads overnight, and the website sanwan.ai was designed, coded, and launched by Sanwan itself.

His sentiment: "No human employee responds immediately, but this one does; you tell it, it acts at once."

Even Goldman Sachs' Philip Sun is amazed; the young Chinese investors he interacts with already have several "lobsters" each:

One summarizes daily market news, one analyzes whether investment decisions are too emotional, and there’s even a "foreman lobster" that oversees the work of other lobsters.

He describes it as: Fascinating.

"Lobster-Raising" Craze: From the Programmer Community to the Elderly

Chinese people gave this phenomenon a down-to-earth name: "raising lobsters".

Beijing, Shenzhen, Hangzhou… from online to offline, OpenClaw sharing sessions and installation events are mushrooming everywhere.

At an OpenClaw meetup organized in Hangzhou, so many people showed up it had to be limited; a follow-up meetup in Shenzhen saw over 900 registrations. Some sell installation services on Xiaohongshu and WeChat, prices ranging from tens to hundreds of yuan, yet supply still falls short of demand.

What’s really shocking to foreigners is how this craze breaks demographic barriers.

On Reddit across the ocean, a user said his grandfather generates AI greeting images every morning to send to the family group. Another user pointed out: "You don’t realize how accepted AI is in China."

Some joked: "If it’s a free event, seniors will definitely go, even if they have no idea what it's about."

US netizen sean_hash commented:

"From announcement to mass installation event in Shenzhen, it was only 72 hours. Such deployment speed is hard to replicate elsewhere."

Some netizens, after watching videos of Shenzhen lit up at night and people nationwide engaged in AI, felt defeated:

"As an American, look at Shenzhen, then look at Baltimore—it feels like we’re living in two different centuries."

A Vietnamese netizen said Asia’s attitude towards AI is generally much more positive, and seeing China’s enthusiasm, everyone else is basically following suit.

Someone even posted: "Should I start teaching my daughter Chinese?"

Why Is It China?

It looks like a nationwide carnival on the surface, but there’s actually a clear logical chain behind it.

First, the public narrative is different.

In China, AI is consistently defined as an economic growth opportunity, written into five-year plans, and appears in the Premier’s government work report—at this year’s Two Sessions, the report explicitly called for accelerating the application of new-generation smart terminals and AI agents, and encouraged large-scale commercial use of AI in key fields. Observers note China’s public narrative about AI is highly unified; in contrast, internal U.S. discourse is more fragmented.

Western observers analyze: Americans and Europeans grew up watching 'Terminator' and 'The Matrix'; fear of machines overthrowing humans is ingrained in their cultural DNA. Chinese people, having experienced firsthand how industrialization and urbanization lifted their living standards, instinctively react to new technology as "it can make my life better".

Second, low cost and low barrier.

OpenClaw is open source, and Chinese cloud providers have made deployment services "one-click installation". Alibaba, ByteDance, and Baidu rolled out foolproof configuration schemes, and Tencent employees set up stalls to help people install it. Coupled with high-performance domestic AI models like Tongyi Qianwen, DeepSeek, etc., ordinary users don’t need to know code—just scan a QR code to start "raising lobsters".

From GitHub codebase to mini programs, living logic trumps technical logic in a dimensional crush.

Third, unmatched integration speed across platform ecosystems.

WeCom, DingTalk, Feishu, Taobao—major platforms almost simultaneously integrated OpenClaw architecture. This means "lobster-raising" isn’t tinkering in an isolated tool but embedded directly into the daily entry points for hundreds of millions of people’s work and life.

Tencent is reportedly developing an AI agent specifically connected to WeChat mini-programs, directly competing with Tongyi Qianwen and Doubao.

Fourth, local governments offer large-scale subsidies to "lobster-raisers".

According to a WallstreetCN article, cities like Shenzhen, Suzhou, and Hefei launched subsidy policies for OpenClaw projects within days of the craze, building industrial parks and issuing grants to "one-person companies".

A technology that travels from programmer circles to government documents in a few days—this speed is unimaginable anywhere else in the world.

Not the First, and Certainly Not the Last

From shared bikes to mobile payments, from livestream commerce to today’s "lobster-raising", the standard template for China’s tech ecosystem has never changed: not necessarily the first to invent, but definitely the first to scale, land, and commercialize fastest and hardest.

Academics call this "second-mover advantage"—no need to burn money and learn the hard way; wait for the tech to mature, then amplify it with stronger execution and a more complete ecosystem, turning it into infrastructure that changes billions of lives.

In the West, OpenClaw is mostly a “cool toy” among geeks, but in China it has already become “national infrastructure” used by retired engineers, housewives and students alike.

Goldman Sachs’ sales are amazed, Reddit netizens are dumbfounded, Jensen Huang says it’s "the most important software release ever", but few expected that the ones most enthusiastically raising lobsters would be ordinary Chinese people.

The highest realm of technology has never been isolated in the laboratory—but is about entering daily life and changing billions of ordinary days.

This "lobster" storm is just beginning, and China’s speed and passion have already made the world hear the sound of the future rushing in.

Risk disclaimer and termsThe market is risky; investments require caution. This article does not constitute personal investment advice and does not take into account the individual investment objectives, financial situation or needs of specific users. Users should consider whether any opinions, views or conclusions in this article fit their own circumstances. Invest accordingly and responsibility is your own.