Manus became popular overnight, but I want to pour some cold water on it

Overnight, Manus became popular. After several AI circle big names enthusiastically forwarded it, countless people circulated it in various groups.

It can be said that many hands make light work. With the atmosphere of the world's first and most powerful general-purpose AI " made in China ", it is even more difficult to get a code. Even the official media account said without any verification that the number of invitation codes has reached 88,000 each.

However, I myself was not that excited.

Here I would like to state my identity as a user: Devin 350 ACUs consumer.

Although I haven't queued up for Manus yet, I am familiar with the use of Devin. One ACU is equivalent to about 15 minutes of Devin work, which corresponds to 2$, which is not much different from the price of one task mentioned by Manus.

My first impression after seeing Manus' showcase was, isn't this just a thing that makes Devin available to people other than programmers?

I have tried many of the cases shown in this article a long time ago using Devin in different ways.

So I think it takes some courage for anyone who has used Devin to praise Manus with exclamation points like " first-of-its-kind ", " awesome ", " crushing ", and " belongs to China ".

Please note that I am not saying that Manus copied and borrowed without any innovation. On the contrary, many great products need a simple and easy-to-understand sentence like " It is xx of xx " to anchor its precise positioning.

Manus does a great job, completely exceeding the expectations of ordinary users, and even gives people a sense of cyber bull and horse - of course, this kind of surprise is only for ordinary users.

The Manus  team should have done several things right (not only) :

1. Be an early follower and user of Devin

2. The idea: "What if devin is expanded from programming scenarios to general user scenarios?"

3. Do it when you think of it

4. Become a deep user of Devin

5. In-depth research and successful implementation or integration of computer use, virtual machines, planning, artifacts and other capabilities

6. It's done, and the effect is pretty good. Although there are still occasional glitches, it basically meets expectations.

What was said above may seem simple, but each step can stop at least 90% of people.

Although the current Manus will make people who don’t understand it not know how to use it, and those who do understand it may not bother to use it (to do real work), it does a good job of implementing the interaction of “delivering results” that Devin first (?) claimed.

And this new form of interaction, after being practiced by a large number of users, is likely to produce several hit scenes - and this is not the final version. After several iterations and upgrades, it is possible that it will really "explode", but it is obviously overrated now.

As for the cost issue of large amount of token consumption that some people criticize, I don’t agree with it.

As long as there is demand, the hardware performance soaring, the price dropping + the model's per-parameter performance (ok, this is an indicator I created) skyrocketing is not a problem at all.

What's more, this is what is happening now.

I believe that soon, a personal PC worth tens of thousands of yuan will be able to run Manus as a local model without Internet access.

No, you need to be connected to the Internet, otherwise how can I check the information and report the results to you?

Here is a quote from a rather sarcastic comment:

So it’s not surprising that it can go viral, after all, OpenAI’s operator is blocked to us. But as to whether it is really that awesome, more professional evaluation is needed to verify it, and it cannot rely on self-indulgent marketing methods.

The marketing methods of "Manus" this time are negative points rolling operation. The first is the hunger marketing of invitation codes, the second is the subversive carnival evaluation, and the third is to anchor the friendly companies to take advantage of the situation, such as this article "China's AI Twin Stars: DeepSeek's technological breakthrough, Manus's commercial landing", and "DeepSeek was beaten? Another domestic AI is popular". No wonder everyone says that its popularity this time is all due to self-media, not professional technicians.

Smart Brain, WeChat Official Account: Atmosphere DeepSeek's "illusion" and Manus's negative marketing

I'm not going to be venomous, but I do want to pour some cold water on you.

What I want to say is that although things are good, when a person, a group, or a circle becomes easily excited about something, it may be time for us to calm down and think independently.

My advice is that if you find yourself so excited that you can't sleep at night, then maybe what you need most at this time is to pour a bucket of cold water on yourself.

Clear your mind and think about the following questions:

What exactly was I excited about? — Was it because I read a passionate article, or did I really feel that change was about to happen?

Why does this product excite me? Is it because it is made in China, is it because everyone says it is good, or is it because I really feel AGI from it?

What is the reason behind Manus's explosion in popularity? Is it starvation marketing, or the guidance of a few "opinion leaders"? Is it the recognition of a large number of users after evaluation, or a complete transformation of the world?

What did my excitement bring me? — Is it the sense of accomplishment of forwarding it before others, or the conversation topic at lunch showing off my knowledge? Or is it that I have gained something from it, grown, become better and stronger? Or is it that it validates your judgment, the project you want/have/are executing? Or is it that as a project member, your efforts are finally recognized by others at this moment?

And Manus's popularity also gave me some inspiration——

What I have always thought is that although  AI’s capabilities are becoming stronger and stronger and can be used to solve many problems, people’s needs have not increased, and people have not had more time. This seems to be a zero-sum game in a stock market.

Even negative-sum games — AI is eliminating some people’s jobs.

How to use AI to find incremental things?

This is something I’ve been thinking about.

Apart from reducing costs, increasing efficiency and having fun, what else is there?

The answer may be AI Native products, products that can open up new markets.

Kai-Fu Lee once gave a side description of AI Native - without AI, this thing would not be true and could not be done.

Manus is clearly AI Native, and has also found an incremental market - AI self-media bloggers and readers.

Oh yeah, and investors too.

This is not irony, but a product that is ahead of demand. These are the seed users. If these people do not buy it, there may be no future.

This is innovation for the future.

Finally, back to the (positive) statement that “Isn’t it just a thing that makes Devin available to people other than programmers?”

I want you to be quiet for a moment and think about this:

Why did they think of it and I didn't?

Why did I think of it but didn’t do it?

Why am I trying this but not succeeding?

Even - why did I think of it and do it, but no one knows about it?

Also, have I learned any product promotion skills from this incident?

Also, I guess that the Devin team must have thought of such a use case, and either they are working on it or they think it is unnecessary. But Devin has not been released yet, what is the possible reason?

And finally, are there any techniques for innovation?

The answers to the above questions vary from person to person, and I don’t have a standard answer.

To be honest, I don’t have much successful experience.

But at least, you can read more first-hand information, less things that are chewed over by others, more dry and cold information, and less texts that are too emotional.

In this way, you won't be startled by things that shouldn't be startled , and you won't be easily ignited and explode and waste your good time in passion.

I hope that you who have read this article can calm down and think about it, so that Manus can be useful to you, rather than you being useful to Manus or self-media bloggers.

As for the last question, I can share a little trick I learned (no nonsense, show me the prompt):

The case may not be very good, so just stop here.