It is meaningless to worry about what ChatGPT will replace

If you care about technology, you should have read several articles about AIGC by now; if you don't care about technology, you probably won't click on this one. Without wasting time, let's talk about a question right now: what is the difference between a human and a machine when it comes to creation?

Creator's intent

If machines can draw faster, write more efficiently, and program with fewer errors than humans, do we need to continue to create as humans? Wouldn't it be better to leave it to machines to create new pictures, articles, poems, programs? It’s also a cliché, a common narrative: machines are created to replace humans in some way. "Replacement" is a neutral concept. If a machine replaces a human, no one will be robbed of their job. But if a computer can produce drawings faster than a painter, and write a manuscript faster than a reporter, it is a problem of another dimension. .

People, no matter who they are, will long to create some marks of their own in their lives. We reflect what kind of existence we are through creation, even a circle of friends must be different. So when the current machines can draw and write, are humans still special? If the machine can think of the title of my article better than itself, what use is it for me?

Even if machines are more powerful than they are now, humans as creators still have the most unique value: intentionality, a strong intentionality . That is to say, as a creator, there must be a goal, hope, fantasy or even fantasy in his mind – but he/she is trying to use creation to reach that other shore.

And this is precisely what machines are most lacking in: they have no intention. Even if machines (at the current level) destroy human beings one day, it will most likely not be intentional. An AI's work expresses what it has learned, not what it intends to express. The AIGC can paint a fantastic work, but it has no "intent" to paint this work out of thin air. There must be a person who has the "intention" to let the machine paint, and this work can be born.

That might sound abstract, so let me give you an example. This is one of the few enlightenment moments in my life:

You already know how to take pictures, so why do you want to take pictures?

To be honest, I still don't know the answer to this question. So I decided to give up photography as an art for the time being, and just treat it as a hobby of daily photography. The person who asked this question was an artist named Tianxi. It was 2019, and I was working on an art project to shoot Zhongguancun, but after four years of shooting, I felt that anyone could take the photos I took.

Tianxi's problems were the last straw. How to find a robot that walks in Zhongguancun every day, and can have similar photos. So why should I shoot it? I don't have a very strong intention in this creation. If it was only about "shooting Zhongguancun", I could be replaced by a machine. I don't have a violent intention in this matter of having to compel myself to create. This intention is a certain kind of artistic determination. An artist friend said that Siken is a kind of creative impulse of "I would rather be hungry than make art", and the intention to achieve the goal no matter what. Some people also call it the creator's "madness".

But there are some things I have strong intentions about. Like this article. I hope to discuss with you what is special about human creation in the age of AIGC. That's what I'm writing for. And if we ask "What is the difference between human creation and AI?", ChatGPT will give this answer:

There are many differences between human creations and AI creations. The most notable difference is that human creation is based on the sensibility and imagination of the human creator, while AI creation is based on the training and data it receives. In addition, human creators can think about problems from many different perspectives and create according to their own personal preferences, while AI creators cannot.

In addition, the creative process of humans is often more uncertain and vague, while that of AI is more precise and predictable. Humans' creative process may go through many iterations and twists and turns, while AI's creative process is more linear and sequential.

In short, human creation and AI creation have many different characteristics, but they can both bring us great value.

This answer is not wrong, but it serves no purpose. AI is just answering the question, it doesn't have the enthusiasm to explore the question and the desire to convince the reader.

But the absence of intent does not mean that the AIGC is unimportant. On the contrary, if you have the intention, AIGC should be a good assistant to help you create. If AI can quickly help me produce a draft of an article, I will not refuse. Because as a writer, what kind of content should I need.

human taste

For creators, there are two particularly important abilities in most cases: the ability to collect data, and the ability to make connections and build new things from it. Even an artist cannot create without the context of art history; on the contrary, any great artist must have seen enough works of art, and it is impossible to create something out of nothing. The skill of "until others have done it" has always been very important in creation.

But after the birth of the Internet, the ability to collect data has been raised higher and higher. The vast majority of creation has become a pile of collected materials-when you can find information in one day that a person 100 years ago could not obtain in a lifetime, it is difficult to suppress this thirst for information. Using articles as an example, a common way of creation is to collect a large amount of information and extract biographies. Countless content is telling you what other people are doing, like the kind of "Top 100 Movies", "Top 10 Must-Reads in Life".

In the age of AIGC, this matter will only be more prominent: because machines are more capable of gathering than humans. This calls for another kind of ability that we also need to pay attention to, the ability to make connections and create new things from data. We often use another name to describe this ability: taste. How creators use these AI creations is ultimately a reflection of taste. How to filter, organize, and collaborate with AI depends not on more information, but on your taste.

▲ Picture from: "Ready Player One"

Today, the word "taste" has a hypocritical negative feeling. If I want to say that creators should be more hypocritical than ever. We can already use almost unlimited information, and AI can help us create faster than before. Shouldn't we look for a taste that is more in line with our own aesthetics? The nourishment of creation is extremely rich, and any taste has its own promised land. The help of AIGC can open the eyes of the creators, and even cover the eyes of the creators. The eyes that are covered are the eyes with your taste.

Beyond that, the AIGC is forcing us to rethink our definition of creation. This type of technology forces us to ask ourselves: Is everything I do really creating? Or is it masquerading as creation but just a bit of mental labor? The history of mankind is the history of the continuous appearance and disappearance of occupations, and no one can promise eternity. And we often mistake our temporary occupations as a natural existence in society because of our several years or more than ten years of experience.

In 1926, the French Jesuit missionaries of the Apostolic Vicariate of Xianxian County, Hebei Province published a pamphlet entitled "Missionary Builders: Proposal-Scheme". The purpose of this booklet is to help missionaries in China build a good church that is not too ugly with as little money as possible. A hundred years later, this handbook was rediscovered and has become an important material for studying the evolution of Chinese Christian architecture and how Westerners work in China. Almost all the operation methods in this manual have been eliminated by time. That is, today, no one or any profession will follow the operation of this manual to construct.

But the passionate intentions of these missionaries are on full display in this handbook. Using this booklet, my country and Germany rebuilt the Daimyo Catholic Church. And this church has been evaluated like this: "The Daming Catholic Church in 2014 is like the Dunhuang scripture cave in 1900. The moment the wall was broken, an academic era has quietly opened."

Just as today there is no need to paint like Wang Ximeng of the Northern Song Academy of Painting, but the green landscape of "A Thousand Miles of Rivers and Mountains" still shows the creative ambition of this painter who has not many records. Careers are fleeting, but the intention to create is forever.

unpredictable future

But what occupations can AIGC make disappear, and what occupations will be born? Frankly, I don't know.

The Zhongguancun art project mentioned above was started because I started my business here in 2015. In a broad sense, our company is doing artificial intelligence – but this word is too big, and the direction of subdivision is computer vision using deep learning. My job is to teach computers to recognize all kinds of things through vision: from the hidden cracks of solar panels, to the moving planes at the airport, and finally to a fast-moving North China in the forest area of ​​Shanxi, only the tip of its tail can be seen Leopard. Later, not only visual images were studied, but also binocular sensors, lidar and a certain technology called SLAM. Anyway, net is some "big words" that the entrepreneurial circle is willing to say. I say this because if talking about artificial intelligence requires a certain qualification, I still have it.

And this is another moment that made me suddenly realize: There were a few years when I was bombarded with massive artificial intelligence news every day, my job was to discuss this with people every day, and every week I listened to various entrepreneurs Talk to investors about the future of artificial intelligence. Today, many years later, I find that almost all the high-spirited talk at that time was bullshit.

No one thought that the artificial intelligence of the future would give you a piece of code trying to destroy the world while chatting, no one said that the computer would help you draw pictures in the future, and no one thought that the breaking of artificial intelligence would turn your photos into a two-dimensional style on this matter. nobody. So predicting the future is a very dangerous thing, because the future is really too random. This may also be the beauty of the word "future", because human beings often do not know what kind of flowers will bloom from the seeds planted. The road to hell may be paved with good intentions, but the journey of redemption often begins with despair.

Don't predict, create. This time with AI.

Contributing author: Wang Hanyang (anchor of "Let's Talk" and "Monster Shangzhi")

Ai Faner makes the future within reach.

#Welcome to pay attention to Aifaner's official WeChat public account: Aifaner (WeChat ID: ifanr), more exciting content will be presented to you as soon as possible.

Ai Faner | Original Link · View Comments · Sina Weibo