After a weekend, the heat of "Lobster" has not only failed to fade but has entered a stage of magical realism.
Even more magical is that the popularity of Lobster in China has even surpassed that of the United States.
Lobsters on cloud hosts, Lobsters on local PCs, Lobsters on mobile systems, and even some Cowork-type products that aren't strictly Lobsters have started putting the Lobster hat on their heads.
Cloud vendors are soaring, with coding plan sales for Volcengine, Alibaba, and Tencent skyrocketing.
Mac minis are sold out, with spot prices doubling.
Figures from the Internet 1.0 era, like Zhou Hongyi and Fu Sheng, have also jumped in to "raise Lobsters."
Overnight, everyone is talking about Lobster, everyone is installing Lobster, and everyone is afraid of not having a Lobster.
Zhu Xiaohu lamented. He said:
"Lobster" is indeed a very obvious starting point, equivalent to the basic determination of the OS operating system in the AI era.
This statement is a bit certain, but it's not without reason.
Lobster is like a "unified access layer."
It connects to models above, terminals below, and corporate IMs and workflows on the side. So it looks like a tool, but it's actually more like an operating system for the AI era.
Many people, immediately after installation, ask the soul-searching question:
"What can I do now that I've installed Lobster?"
This question is very interesting. Everyone is installing Lobster, but no one knows what to do after finishing.
This point makes it even more like an operating system, just like when you open a computer and install Windows—it seems like you can do anything, but you really don't know what to do.
That's why Microsoft later bundled IE and Office into Windows.
Today's Lobster is also in this new system stage.
Short-term Bubble or Long-term Dividend?
Is Lobster a short-term craze or a long-term opportunity?
A viral thing is often short-term, but its impact can be very long-term.
The government's reaction speed to the Lobster matter has been surprisingly fast.
Shenzhen Longgang took the lead in launching the "Lobster 10 Measures," and Wuxi followed with the "Lobster 12 Measures." Behind this is the government vigorously promoting the implementation of OPC policies.
Shenzhen Longgang even announced it will co-host a thousand-person Lobster conference with Kimi, with official involvement driving industry development.
OPC policy, the One-Person Company, is good for the country and the people.
Previously, encouraging mass entrepreneurship required building teams and getting investment, which had a high threshold.
But with OPC, one workstation + a little subsidy + countless AI Agents truly allows one person to become an army.
The government's rapid response is an attitude, giving people a sense that it's worth long-term investment.
Additionally, an old article has been circulating in various groups. The trivia is that most people only look at the title and not the content, so everyone mistook that lobster for this Lobster.
Who would have thought of such a bizarre propagation path?
But behind this propagation is people's positive and optimistic attitude toward this matter.
Are They "Installing Lobster" or just "Posing"?
For something to truly go viral, it usually relies on only two things: emotional value and commercial value.
Emotional value is responsible for spreading; commercial value is responsible for survival.
With the DeepSeek wave, the rise of Chinese open source provided plenty of emotional value; cloud vendors and ecosystem service providers reaped the commercial value.
And this time with the OpenClaw Lobster craze, emotional value and commercial value exploded almost simultaneously.
Raising Lobsters itself carries extremely strong emotional value.
The popularity of Lobster has fueled the on-site installation service business.
Charging 500 to over 1,000 RMB per installation has become the fastest-starting side hustle direction currently.
The most absurd scene is an installation master saying:
"I've installed many Lobsters recently; demand is quite strong, although I myself don't know what they can do after installation."
This sentence is perfect.
What can you do after installing it? This really is a soul-searching question.
An installation master can help you install Lobster, but they can't help you figure out what Lobster can do.
Previously, the motivation for users to install a new product was "I know what it can do!" Now, it's regardless of anything, install it first, and then ask, "What on earth can I do with this thing?"
We provided ListenHub's Lobster Skills to users, but users came to ask us if ListenHub has a Lobster installation service...
It's truly... upside down.
You might not know how to use it, but you can't not have it.
You might temporarily not know what to do, but you must install it first.
Whether you've installed it or not is no longer just a tool issue, but a matter of posture—an identity issue of whether you've kept up with this era.
Can You Really Make Money Raising Lobsters?
After talking about emotional value, let's look at the more critical commercial value.
Lobster isn't just noise this time; people are actually starting to make money.
First to make money are the model companies.
A few days ago, my friend Gefei posted an important data point on Xiaohongshu: In January 2026, Kimi's orders on Stripe surged by 8280%, and grew another 123.8% in February. In the 24 months from 2024 to February 2025, Kimi's orders never entered the Stripe top 100, but in January they soared to 22nd, and in February to 9th.
Entering 2026, Kimi's cumulative revenue in nearly 20 days has exceeded the total for 2025. From then on, no one doubts Kimi's commercial potential, and the company's valuation has surpassed the $10 billion threshold.
But it's not just model companies making money in this wave.
If we look closely at this chart, we can find that not only model companies like Kimi and MiniMax are making money.
Manus (acquired by Meta), Alibaba's Copaw, and startup HappyCapy have all benefited from the Lobster traffic dividend.
The most surprising thing is the huge shift in the attitude of big tech companies.
Feishu—the whole team is particularly obsessed with Lobster. Feishu's previous robot access process was practically anti-human, but the official team first made a set of plugins to reduce friction, and today they are aggressively giving everyone a Lobster directly.
WeChat Work also quickly announced support for OpenClaw access.
And supporting OpenClaw means supporting all Lobsters.
Yesterday, Da Congming even wrote a Kimi Claw WeChat Work Access Tutorial:
The significance of this is very major.
Previously, if a new company's product wanted to break into enterprise service IMs, it was extremely difficult—closed ecosystems, page blocks, account penalties.
But now, through this shell of Lobster, you can directly enter the interior of various enterprises. No need for employees to install separately, no need for IT departments to deploy separately, no need for your sales team to say a word.
Kimi itself never thought it could enter thousands of enterprises in such a way.
And why exactly is WeChat Work breaking barriers and opening its ecosystem?
In the end, it's all about one word: "Money."
Everyone has talked about enterprise service AI for years, but none has truly made enterprises spend money on a large scale.
Lobster isn't a platform educating users, or a vendor convincing enterprises, or sales pushing hard. Instead, users themselves rush to the door saying: I want to install, I want to connect, I want to buy.
There is no commercial signal more moving than users actively wanting to buy.
For common interests, humans can eliminate all barriers.
The whole world hustles for profit; the whole world bustles for gain.
Everyone who makes money with Lobster becomes a behind-the-scenes promoter of the Lobster craze.
Behind the Frenzy, Calm Reflection
Lobster's madness is real madness, but I've looked into it and think it might not be a bad thing.
First, the world is truly becoming a bit reconnected.
This time, everyone is not acting on their own.
Previously, you pushed Doubao, I pushed Qianwen, he pushed Kimi—everyone was building their own walled gardens.
But this time is different. WeChat, QQ, and Weibo have all connected to Kimi, and then Kimi connected to OpenClaw.
More importantly, the enterprise service barriers built by various platforms are now being self-dismantled.
Feishu, DingTalk, QQ, WeChat Work—how many years and how much money did these platforms spend to build their moats? Now they are actively opening them.
This feels very much like that long-lost feeling from the Internet 1.0 era:
Connect first, talk later.
Second, Lobster can indeed make money, and users are indeed willing to spend.
Cloud vendor sales are skyrocketing, AI model sales are skyrocketing, hardware vendors' inventories are cleared, and even installation masters can earn over 10,000 a month.
It's not about grinding for free or giving away eggs with subsidies, but real money-making.
Users have a strong willingness to pay; enterprises have a strong willingness to pay.
Before they even know what value it can bring, they are willing to open their wallets.
Chinese people have money now, and their mindset is that of wealthy people.
The stereotype that Chinese people don't like to pay for software needs to be completely changed.
Third, not knowing what to do means the ecosystem is still empty.
Many people see "everyone installing Lobster but not knowing what to do" and their first reaction is: Bubble.
But in my view, this precisely shows that the upper-level ecosystem of the Agent OS is still a blank space.
If a system is already so mature that everyone knows how to use it, there aren't many opportunities left.
It's precisely because everyone is still asking "what can it do" that it shows the real application layer, scenario layer, and service layer haven't been filled yet.
Emptiness is not a bad thing; emptiness means opportunity.
Whoever can find their position and create their value in this empty ecosystem will make money.
Remember the sentence I wrote in an article:
"Agents are the demographic dividend of the new era."
Looking at this sentence now, it feels even more relevant.
With so many people across the country having installed Lobster and waiting to be fed, the words "demographic dividend" are basically written on their faces.
After the Fever, What Should We Do?
The popularization of something inevitably comes with frenzy, misunderstanding, bubbles, and chaos.
This is the standard process for humans to accept something new; it's not embarrassing, and it's very normal.
What matters is not whether the process is elegant enough, but that it has already spread.
Lobster has become a new operating system. The world has reconnected.
Lobster can indeed make money, proving that this path is open.
Although the ecosystem is currently empty, it gives us the opportunity to build.
The important thing remains:
Think about what we can do in this, or what we can do to help Agents.
Not following the crowd to install Lobster, but finding your own position on this new operating system.
Not chasing hot spots, but creating value.
Not doing what everyone else is doing, but doing what only you can do.
The people who will truly reap the dividends in the future are not necessarily the first to install Lobster,
But the first to answer that question:
After installing Lobster, what can you actually do?
