If we were to select the most fascinating new technology in 2023, it would be ChatGPT. From Silicon Valley and Wall Street in the United States to Houchangcun and Zhongguancun in China, almost all technology giants are discussing “generativeAIAfter ChatGPT, new generative AI products were released almost every week. Large models such as Wenxin Yiyan, Tongyi Qianwen, and Gemini have emerged one after another. In the sub-sectors, Midjourney, Miaoya Camera, HeyGen, and Pika have become popular overnight...
No one dares to ignore this technology. Zhang Yong, then chairman and CEO of Alibaba Group, said at this year's Alibaba Cloud Summit that all industries are worthy of being based onAIRe-do the technology. In May this year, Baidu CEO Robin Li expressed his opinion that "big models will change the world" and "Baidu will use big models to re-do all products." Internet giants such as Tencent, ByteDance, Kuaishou, Didi, and Pinduoduo have also followed suit in the AI business.
The development of technology is often the innovation of productivity tools. On the one hand, AI has spawned more new occupations, such as AIGC painters, prompt engineers, etc., and the large gap and high salary have attracted attention. According to relevant data from Liepin, the demand for artificial intelligence talents is nearly 3 times that of five years ago, and the Internet as a whole is 1.06 times. From the perspective of the overall demand for AIGC (artificial intelligence generated content), from January to August 2023, the number of new AIGC positions increased by 139.76% year-on-year.
On the other hand, the popularity of AI has also caused some people in the workplace to feel anxious about their jobs being replaced by AI. As a result, those who are unwilling to be replaced are willing to spend a lot of money to buy courses and learn AI. However, the jobs of some people who cannot keep up have already been replaced by AI.
The emergence of any new technology has two sides, and the current hot AI is no exception. No one can stop the advancement of technology. For ordinary people, AI, as a new technology, can both make them successful and affect their work. The key is to learn how to use technology better, rather than being replaced by it.
Taking advantage of AI, earning 30,000 yuan a month
He Lihua's income suddenly tripled, from 10,000 yuan a month to 30,000 yuan a month, which was something he could never imagine as an ordinary undergraduate graduate.
He Lihua was very worried when he chose artificial intelligence as his major. Compared with computer science, software engineering and other majors, artificial intelligence lacks a broader job market. Even if there is room for employment, artificial intelligence companies require graduates from 985 or 211 universities, which is a high threshold. So after graduation, he chose to support himself first and seek stability.
The turning point occurred at the end of last year. In November 2022, the chatbot ChatGPT released by the startup OpenAI attracted 1 million registered users in just 5 days, becoming one of the fastest-growing products in history. Open AI has become the world's most valuableHighestAIGC is a global startup.
This gave He Lihua hope. He Lihua felt that domestic AI entrepreneurship was still in its early stages, and choosing a suitable AI company to work for might bring broader prospects.
He Lihua is currently an AI trainer. Compared with AI labelers who can just use computers, his position requires higher technical skills. He Lihua is responsible for collecting data from various sources and preprocessing and cleaning the data. He also develops and debugs models.
Artificial intelligence trainer is a new profession that has emerged with the development of AI. There are also similar positions such as prompt word engineer, which are highly professional positions. For example, the large model data annotation base established by Baidu in Hainan requires applicants to have a bachelor's degree. Of course, AI has also spawned some new professions that ordinary people can hold, such as AI cartographers and AI narrators.
As it turned out, his choice was not wrong.
There are few technologies that make almost all Internet giants at home and abroad excited. At the end of the same year, ByteDance recalled Zhu Wenjia, the technical director of TIK TOK, and launched a conversational AI project code-named "O" internally. Two months after the release of ChatGPT, Baidu, which has been betting on AI for many years, said at the Creat AI Developer Conference that it will accelerate the integration of AIGC and search to improve the search experience.
The AIGC wave is rolling in. Bill Gates, co-founder of Microsoft, is certain that "this is the most revolutionary opportunity in 40 years"; Nvidia CEO Huang Renxun exclaimed that "the iPhone moment of artificial intelligence has arrived"; 360 Group CEO Zhou Hongyi asserted that AI big models are industrial revolution-level productivity. Baidu CEO Robin Li believes that "we are at a new starting point. This is a new era of artificial intelligence with big models at its core. Big models have changed artificial intelligence, and big models will change the world."
superInternet companies flocked to the market. In March this year, Baidu took the lead in releasing Wenxin Yiyan, followed by Alibaba, iFlytek, 360, Huawei, JD.com, SenseTime, etc. According to data from the Beijing Municipal Bureau of Economy and Information Technology, as of October 2023, the number of large models publicly available in China alone has reached 238.
Powerful players all hope to create a universal large model, just like there are two operating systems in the mobile phone era. They hope that their models can also grow a prosperous ecosystem. Who can be the fastest?mostWhoever releases a general large model will have a greater chance of becoming the underlying operating system.
They want to be reviewed as soon as possible. As a result, Baidu's Wenxin Yiyan, Byte's Yunque, Alibaba's Tongyi Qianwen and other large models have appeared one after another.
Benefiting from the emergence of these large-scale products, some people have begun to use these products and tools to create new ways to make money for themselves. Novel writer Lin Tong is one of them. She found that AI may replace human labor to automatically generate novels. So Lin Tong changed the traditional writing mode and began to use Wen Xin Yi Yan to write.
"What I have to do every day is to think of a story background, and then break down the story into details. The details must be broken down detailed enough so that the text feedback from the AI can be more detailed and real. If the questions are asked too broadly, the feedback will be empty and not fascinating," Lin Tong shared her experience with Tech Planet.
Lin Tong said that unlike online plagiarism software or platforms, articles created by AI tools are more unique. Therefore, if you publish "original" stories on the Internet or some text creation platforms, you will have the opportunity to receive incentive bonuses from the platform.
Lin Tong said that emotional, suspenseful, situational and other types of short stories created through AI tools can indeed be fascinating. One emotional article has been read over 30,000 times on the WeChat public account, which is considered to be one of the better ones written recently. The current daily income is estimated to be more than 200 to 300 yuan.
However, risks also occur from time to time. In November this year, the first AI copyright case was sentenced and 500 yuan was awarded. The reason is that on February 24, 2023, the plaintiff in this case used the open source software Stable Diffusion to generate a picture by inputting a prompt word, and then published the picture on the Xiaohongshu platform under the name "Spring Breeze Brings Tenderness". Later, the plaintiff discovered that a Baijiahao account used the picture in question when publishing an article without obtaining his permission, and cut off his signature watermark on the Xiaohongshu platform. For this reason, the plaintiff brought the defendant to court.
Lin Tong also told Tech Planet that although tools can improve work efficiency, they can also cause trouble and risks. A friend he knows was judged to have plagiarized by the platform because he relied on AI tools to write stories. Later, he learned that it was the AI tools on the Internet that re-washed other people's stories.
The replaced workers
In May this year, technology giant IBM announced that it would suspend recruitment for positions that can be filled by artificial intelligence. Prior to this, Meta, Microsoft, Amazon and other technology companies had stopped recruiting for AI positions, and they all said they would adjust their recruitment plans to follow the footsteps of AI.
The reason is that AI has been increasingly used in many fields such as automated customer service, writing texts, art design, and generating code. IBM CEO Arvind Krishna said in an interview with the media, "I can easily see that in the next five years, 30%'s jobs will be replaced by artificial intelligence and automation."
What Arvind Krishna said is not alarmist. In China, there are also some jobs that are being replaced by AI or are about to be replaced by AI.
Li Kuo's situation is not optimistic. He works as a product programmer in a large company and has participated in the development of many excellent software and applications. However, the rapid development of AI technology has gradually made him feel that his job is in jeopardy. In order to improve the efficiency of code writing, the team leader began to let them use AI code generators to write low-end and mid-range codes. Technology has brought the "35-year-old crisis" ahead of schedule.
AIGC can automate programming tasks by analyzing large amounts of data and patterns, thereby reducing the workload of manual code writing. For example, previously complex algorithms needed to be manually written to process data, but now these tasks can be completed automatically using just an AI code generator.
Li Kuo told Tech Planet that the process of being replaced by AI code generators happened gradually. At first, AI code generators only assisted him in completing some basic coding tasks, such as code completion and error checking. Now, it has begun to involve more complex programming tasks, such as automatic code generation and algorithm optimization. Slowly, AI code generators can also do well in tasks that were once thought to be only human-capable, and even better than humans in some aspects.
Li Kuo is astonished by the progress of technology. In the past, a project might have required several days of continuous work to complete, but now AI code generators can complete the same amount of work in just a few hours. Li Kuo believes that part of his work is being replaced by AI, and although it cannot be completely replaced yet, this sense of crisis is real.
Some people are not as lucky as Li Kuo. Shi Wei used to work as a content editor at an Internet startup. For Shi Wei, this job was originally a dream because he could share news and information with readers by writing and editing various content, and collaborate with authors and experts in different fields to create an interesting and insightful online platform.
Shi Wei said that content editing is a job that can be easily replaced by AI. More and more generative AI tools are being applied to his industry. These tools can use natural language processing and machine learning technology to generate high-quality content, almost at the level of manual editing. They can automatically find and organize information, analyze and summarize data, and even participate in editing and proofreading. The company has asked them to start using AI tools such as Baidu Wenxin Yiyan to do content-related editing.
Shi Wei began to worry that if AI tools could complete most of the editing work in a short time, would his job still be necessary? But time waits for no one. After a few months, in order to reduce costs, the company finally hired a group of people who could use AI tools to do content editing. His job was replaced by AI editing tools, and he eventually chose to resign. The choice to resign was also simple. AI tools made the work no longer creative and made him a member of the assembly line.
Since August this year, many domestic technology companies or research institutions such as Baidu, SenseTime, and iFLYTEK have registered in accordance with the "Interim Measures for the Management of Generative Artificial Intelligence Services" and officially opened their self-developed AI big models to the public. This marks that domestic big models have begun to move from small-scale testing to large-scale application. At the same time, it also allows the hidden concerns brought about by the wild growth of big model products in the early stage to be regulated and standardized.
However, the process of AI replacing human labor is still accelerating. On November 7, Open AI's first developer conference ignited AI Agent. OpenAI has released the early form of AI Agent product GPTs and launched the corresponding production tool GPT Builder.
On November 13, 2023, Microsoft founder Bill Gates wrote a thousand-word blog post for Agent, saying it would subvert the software industry and the way people interact with computers: "Whoever can dominate the personal assistant Agent will be a big deal. Because you will never go to a search site, a productivity site, or Amazon."
Users can generate their own GPT simply by chatting with GPT Builder and describing the desired GPT function. Exclusive GPT can be more applicable in daily life, specific tasks, work or family. To this end, OpenAI has also opened a large number of new APIs (including vision, image DALL·E3, voice), as well as the newly launched Assistants API, so that developers can more easily develop their own exclusive GPT.
This means that everyone will have their own personal assistant. This will greatly improve the efficiency of human production and lifestyle. In the future, humans may not only free their hands, but may also have a true digital avatar.
Just a few days ago, Google releasedup to dateThe AI model Gemini was tested on a range of thorny questions, including logical fallacies, everyday moral issues, medical problems, economics and geography.firstDefeating humans shocked the industry.
Google said that Gemini can solve a series of problems such as homework problems and real-time video processing, which means that Gemini is not only capable ofStrongest, and the most versatile, and can be adapted to a variety of tasks, including replacing human work.
For those who have not yet adapted to AI and those who are too easily replaced by AI in the workplace, AI is becoming a hidden concern.
Embracing technology is the way forward for the future
More than 200 years ago, the improved steam engine introduced by Watt increased efficiency by three times, allowing humans to enter industrial society faster and greatly improving textile efficiency. More than 100 years ago, German inventor Otto created the world's firstFirstThe first internal combustion engine, trains and cars appeared, and the travel radius of human beings was rapidly expanded. As long as it is used reasonably, the development of technology will always bring good news to people.
As Robin Li said, "Although some jobs have disappeared, more opportunities have emerged. People's improved work efficiency can support more people, and everyone's life has become better than before."
Not only are there more AI-related jobs, but salaries are also rising. According to the 2023 AIGC Talent Trend Report by Maimai Gaopin Talent Think Tank, with the leapfrog development of the AIGC field, AIGC talents have attracted much attention, and talent salaries have reached new heights. Industry giants such as the Internet, e-commerce, artificial intelligence, and enterprise digital services have all offered high salaries to recruit talent, and well-known hunting companies have offered a monthly salary of 100,000 yuan to poach talent. Mid- and high-end AIGC talents have become a "must-fight" for strategists.
Moreover, AI talents are not limited by the "35-year-old crisis" like the Internet industry. The report points out that unlike the Internet industry, nearly 70% of practitioners in the AIGC field are over 30 years old, and more than 34.5% AIGC talents are over 35 years old, which is 6% higher than the Internet industry.
The AIGC industry generally focuses more on technology and innovation rather than age and years of work experience. For ordinary people, the AI industry is more inclusive in terms of age.
In addition, Li Shikang, an HR from a large Internet company, summarized to Tech Planet that nowadays large Internet companies are stepping up their recruitment of AI talents, but they will also help those employed talents who do not have AI knowledge reserves to adapt to AI, cooperate with AI, and improve work efficiency.
History has proven many times that new technology changes will bring pain, but every technological revolution has elevated human life to a new dimension. In this wave of AI technology, it is better to embrace it actively rather than be afraid. Under the many opportunities, AI is a new development direction for ordinary people.
(Note: The names of He Lihua, Li Shikang, Shi Wei, Li Kuo, etc. that appear in this article are all pseudonyms.)