It’s here, the strongest model GPT-4Turbo! When I woke up, ChatGPT completed the update and launched the strongest model and open application store GPTs. I would like to take a rational look at some of the impacts on the programmer industry after the release of GPT-4 Turbo from a programmer's perspective.
Preface
At the OpenAI Developer Conference on November 7, Sam Altman demonstrated many innovations with revolutionary features of GPT-4 Turbo. Compared with the previous GPT-4 Turbo version, this new version is more like an innovation. Generally speaking, it can be divided into the following aspects:
-
A more complete knowledge base and easier document analysis. ChatGpt has updated its knowledge reserve to April 2023 in the update. At the same time, its text input length has also exceeded 128K context (the previous maximum was 32K). In addition, users can also upload any documents, including PDFs, data files, etc. GPT can help you analyze and refine these documents.
-
No longer limited to text analysis. The new GPT version opens the gpt-4-vision API related to visual capabilities. Through this API developers can complete the processing of video-related tasks. As soon as this function was launched, developers were already using it to explain the game between skt and lng.
(Bokakuharaji site:twitter.com/xiaohuggg/s…)
-
GPTs store is online. GPTs are customized versions of ChatGPT. Just like Apple's Apple Store. Now everyone can create their own GPT and publish it for everyone to use. In the process of publishing an application, we can communicate with GPT using natural language, and GPT will automatically build the application according to our needs. This process does not require you to know any programming knowledge! Currently, GPTs has released nearly a thousand applications.
AI is constantly lowering the threshold for programming
Programmers always seem to be a very controversial profession. From the perspective of outsiders, this group of people just sit in the office every day and fiddle around on the keyboard a few times, and the salary they receive may be several times that of other industries. It is undeniable that this phenomenon may indeed have existed in the early years when the Internet became popular. However, with the development of the past few years, the programmer industry has begun to become seriously involved, and the threshold for programmers has also been continuously lowered.
In the past, you might be able to get a job if you knew a little bit about Java basics, but now the difficulty of interviewing is much higher than before. At the same time, terms like "front-end is dead" and "back-end is dead" continue to top the hot searches. Programmers have always been an industry at the forefront of the storm.
Looking back at the changes over the years, it is not difficult to find that from the initial popularity of drag-and-drop low-code platforms to the unprecedented success of ChatGPT, the threshold for ordinary people to get started with programming is constantly lowering.
Now ChatGPT can even generate relevant code based on the prototype design artwork you pass. Furthermore, prototype artwork may be generated through ChatGPT. All you have to do is act as Party A and continuously provide requirements to ChatGPT, and it will improve the product step by step according to your needs.
Programmers seem to be the next industry affected by the technological innovation of ChatGPT. In fact, many industries will be affected by AI. For example, UI design, text editing, etc. I have to admit that AI is actually affecting all walks of life. For this reason, many people have begun to devote themselves to AI entrepreneurship. Because of this, many shell ChatGpts have appeared on the market, but this iteration of GPT-4 Turbo, Undoubtedly, these case products reveal their true colors.
Where should programmers go?
So what can we as programmers do in the face of the menacing ChatGpt? I think the best way to deal with it is to find an area and delve deeper into it. Only when you have a deep understanding of a field can you have your own advantages that laymen cannot match and directly join the wave of industry development. To illustrate this point, the author here gives two simple examples to explain in detail.
Due to work reasons, I sometimes go down to the production line to learn about the entire process of product production in actual production. When you go to the production line, you will come into contact with some "old masters" on the production line. They have a clear understanding of the entire process of production line production, so they can quickly locate problems encountered in production. In other words, they have very rich experience in the field of machine production.
And I am just an executable program that converts the industrial production process into an online one. In other words, I am just a porter of the process. If one day AI begins to mature and become popular, and companies start to use AI programming to realize online monitoring of the production process, I think the old master will definitely be the one left behind, because he is an expert in production line production, and I no.
People know math better than me! That's what I'm trying to convey. In fact, when you really understand a field and new technologies emerge, you will naturally think about how to make new technologies serve your field.
ChatGPT is based on some data from the entire network, and a lot of data comes from open source. ChatGPT is definitely an efficiency improvement first, and will soon bear fruit in all walks of life. However, if you want to develop applications in some small fields, its data is very small, and its generalization may be poor in the initial stage. So I think this area is still another outlet - low code is better.
The low-code development model in which everyone is a "developer" configures and creates application software through a graphical user interface, rather than relying mainly on handwritten code like the traditional model.
The JNPF rapid development platform encapsulates thousands of common classes on the front and back ends for easy expansion; it integrates a code generator to support front-end and front-end business code generation to meet rapid development and improve work efficiency; the framework integrates forms, reports, charts, large screens, etc. Various commonly used demos are easy to use directly; the back-end framework supports Vue2 and Vue3.
Official website:www.jnpfsoft.com/?csdn. If you don’t want to hear my blabbering, you can go directly to the website to experience it. If you have free time, you can do some knowledge expansion.
By now, I believe you understand what I mean. Returning to the programmer industry, if you work every day on CRUD, and then implement it according to the needs of the business connection, you have never really thought about the logic behind the business, and you just use the framework, and you have never conducted in-depth analysis and research on the bottom layer of the framework. Pass. In this way, even if you work for ten years, neither the work content nor the understanding of this programming in your head will be different from ten years ago. So do you think you will be the person who will not be eliminated?
Summary
The change of technology is the general trend. Even if AI does not appear now, BI and CI will eventually appear in the next few years, and new technologies will always emerge in endlessly. Perhaps, whenever a new technology appears, it will always cause you anxiety, but in the final analysis, what really makes you anxious is not the new technology, but the failure to take root in a field. So you don’t understand and have never felt that technology serves the field.
You just feel that when new technologies come, it seems that my job is going to be eliminated. In fact, even without new technologies, these jobs will eventually be eliminated, because human nature is to be lazy and always think of other ways to liberate productivity.
So do you think the arrival of GPT-4 Turbo will make programmers unemployed? I don't think so. You might as well consider this question: Is GPT-4 written by GPT or programmers? In other words, do you think the underlying facilities that OpenAI operates, from hardware facilities, to system architecture, to software development, and data storage, are all designed through AI? I think the answer is definitely no.
Therefore, programmers will not lose their jobs due to the arrival of AI. On the contrary, there may be a higher demand for high-level programs in the future, because AI has lowered the entry threshold for computers, but it has never lowered the upper limit of computers. Just like Springboot for Java developers, it is a tool in itself, which can greatly reduce the difficulty of program development for the web. And now the emergence of ChatGpt has greatly simplified the entry threshold for programs, allowing people who do not learn computers to quickly build applications.
So my point of view is that programmers will not be unemployed. The prerequisite for not losing your job is: you need to take root in the field you are currently in and keep digging deeper.
I hope the article will inspire you to learn. If you think the article is helpful to you, please like and follow it so as not to miss any subsequent updates from the author!