Author | Mr.K Editor | Emma
Source | Technical Leadership (ID: jishulingdaoli)
"Humans have become accustomed to monopolizing complex wisdom and becoming the center of these wisdoms...but the emergence of artificial intelligence will change human beings' perception of themselves." This is a passage from the book "AI Generation and Our Future", the author is Former U.S. Secretary of State Henry Kissinger, former Google CEO Eric Schmidt, and MIT's first dean of the School of Computing Daniel Huttenlocher.
Although we can look at the impact of artificial intelligence on the entire human being from such a high level without the big shots. But we can still feel more and more clearly that the real impact of the AI era on each of us is so real that we can hear the sound of the rice bowl being smashed.
Just a few days ago, the major media exclaimed that "AI has finally started to steal the jobs of humans". After the use of AI at the beginning of the year, 2/3 of the staff has been laid off... This news was reprinted by various citations, causing extensive discussions among countless netizens, and became a hot topic of the day.
Not only that, some manufacturing industries have also begun to test the waters of AI. A shoe-making company with its own brand in Putian began to introduce Stable Diffusion tools into the design and advertising process. The boss said excitedly that the work that required 5 designers to complete in the past now only needs 1 designer. It also mentioned that the efficiency of AI in advertising marketing is unmatched by humans. As long as key data such as input-output ratio are set, AI can make thousands of advertisements and distribute them on appropriate channels. The workload, in exchange for manpower, at least dozens of people need to cooperate to complete it...
In addition, as far as Brother K knows, the foresighted Dachang people also arranged AI early. Ali's recommendation algorithm team has already used ChatGPT to replace search engines in their work, and asked ChatGPT to help write code. People in some game studios of Tencent have also used Midjournry to make pictures intelligently, assisting the planning post, operation post, art team, and UI team to communicate more intuitively.
Regarding AI grabbing jobs, I used to only hear the sound of stairs, but now I can vaguely see someone going downstairs. According to a survey report: " In the next 5 to 10 years, 40% of the people in the workplace will be replaced by AI, and a large part of them are junior and middle-level personnel ."
But isn't it because of this that everyone has to be anxious? In fact, it is unnecessary. Although the generative AI represented by ChatGPT is an extremely powerful tool, humans also have their irreplaceable capabilities and advantages. Jeff Colvin, a well-known reporter for Fortune magazine, once wrote in the book "The Man Who Will Not Be Replaced by Robots":
"Technology is no longer important, and humans are not as good as machines in this regard, and our real advantages come from the deepest human instincts, including empathy, creativity, sense of humor, logic, social sensitivity, and The ability to tell a story. The ability that these machines cannot achieve at the moment is precisely the aspect that needs to be cultivated in the future to make us stand out in the technological world.”
So, when it comes to real and realistic workplaces, who are the least likely to be replaced by menacing AI?
01
people who are good at making complex decisions
Brother K participated in a dinner two days ago. During the dinner, everyone talked about the power of ChatGPT and its possible impact on all walks of life. One of my friends, who is engaged in legal work, poured cold water on everyone. He did not fully agree with some over-exaggerated AI "legends". In his opinion, ChatGPT is difficult to do things that require complex decision-making like his work.
He gave an example, in the handling of various types of disputes, sometimes it is not just a simple application of legal provisions, but also some basic principles of morality and ethics. In the end, experienced judges will make judgments that conform to human nature and do not violate the law.
On the other hand, ChatGPT can only give users some suggestions and information to assist decision-making based on algorithm training and big data. It may still work for some simple logic that can draw a unique conclusion, but for some things that require multi-dimensional thinking, no definite rules, It is difficult to apply to the decision-making of complex events that cannot be clearly quantified. What this friend said gave a lot of inspiration to other people including Brother K at the time.
Herbert Simon, a famous American management expert, once said: Decision-making is the heart of management, management is composed of a series of decisions, and management is decision-making. In actual work and life, when we make a decision in a complex environment, we not only want a certain "conclusion" itself, but also consider whether it matches with many factors such as timing, risk, efficiency, and feasibility.
For example, someone recommends a financial product with good returns to you, and this product seems to have no faults in terms of income model, company background, and management team. If you only analyze from the perspective of known information and data, AI may give you a positive decision-making suggestion, but we, who have been educated time and time again by the investment market, are likely to make completely opposite decisions to AI.
Many problems in reality are often full of complex variables. Decision-making is definitely a systematic process that combines multi-dimensional thinking, multi-angle research, and multi-faceted games. Even if ChatGPT has a huge database and a well-trained data model, at best it can only be a calculation in a digital space, not a test in a physical space. And complex decision-making just needs to be extended to various contacts and interactions with the real physical world. And this is exactly what AI will be difficult to match with humans for a while.
02
people who know how to ask questions
Just as the age of steam widened the power gap and the Internet age widened the gap in information acquisition, the AI era widened the gap in human learning ability. Being able to ask questions is precisely an important manifestation of strong learning ability and logical reasoning ability. After the emergence of ChatGPT, the new position "Prompt Engineer" derived from it is actually a person who specifically asks questions to ChatGPT. He does not need to write code. good smell.
Not only professional Prompt Engineers, if we general public can ask questions better, we don’t have to worry about being replaced by ChatGPT, but we will tune it into an excellent tool to help ourselves improve and grow.
Some time ago, Brother K's company received a promotional advertisement. I asked two young colleagues in the company to try to use ChatGPT. Soon, a little girl came to give feedback, telling Brother K that the output of ChatGPT was completely ineffective, it was just a rough structure full of clichés, with no reference value, and he had to do it himself.
In the afternoon, the plan completed by another little girl through ChatGPT also came out. It was completely different from the fake big empty frame that the first little girl showed me. Her plan was detailed, thoughtful, and executable, which made me doubt it. Is this exactly done with ChatGPT. The little girl told me that this was basically done by ChatGPT, but she made some simple combing and modification at the end.
I asked her again how to use ChatGPT to achieve this effect, and the little girl's answer was just two words: ask questions. Then gave me a brief explanation. After entering the basic requirements, you can also continue to add some detailed questions related to the requirements in the conversation with ChatGPT, such as the main characteristics of the requirements, the goals you want to achieve, and project preparation. The promotion budget, the characteristics and adaptability of various promotion channels, etc. Then ChatGPT will continue to refine and optimize the output plan. The more specific your question is, the more specific its reply will be. It can also continuously give reference marks and links to the original text of relevant content to help you continuously expand your thinking space. After a series of effective questions, the content you get is no longer a simple and rough empty framework, but very close to the final answer you want.
And the girl who first gave me feedback that ChatGPT was useless, gave me a plan she made manually by staying up late the next day, but to be honest, the plan she worked so hard to make is not as good as another girl who completed it with the help of ChatGPT The program is more comprehensive and better.
The same tasks, using the same tools, have the same sense of responsibility, but the final efficiency and results are very different. If Brother K's company needs to optimize layoffs in the future, I will only keep one of the two little girls. Who do you think I will keep?
03
good communicative person
American best-selling author Daniel Pink once said that even the most sophisticated software configured on the world's most powerful computer cannot understand human emotions. The important way for humans to express their feelings is to communicate and socialize. At present, this is still one of the most obvious shortcomings of ChatGPT.
Someone once tried to seek comfort from ChatGPT, asking it "what should I do if I lose my job". ChatGPT immediately gave a reply, "I'm sorry to hear that you are unemployed. Unemployment is caused by many reasons... You must not be pessimistic and disappointed, and you must recognize your ability and value." There are a lot of blah blah blah.
Someone else asked "I feel like my life is meaningless" and ChatGPT's response was "If you feel like your life is meaningless, it might help to reflect on your values, passions, and relationships..." This... such an answer can't It is wrong, but it is indeed not "human words", and the correct one is worthless. In reality, there are friends, colleagues, and customers who ask you for comfort. If you respond to others with this tone and stereotypes, it is estimated that the boat of your friendship is about to capsize.
All ChatGPT's answers are based on cold data and models, without emotion. Through appropriate and subtle interpersonal communication, we can win the business that the other party can give us or not give us in the coffee shop or in the wine bureau; to add to your address book. And such a result is difficult to achieve no matter how powerful ChatGPT is. Because we have human emotions and ways of getting along with humans, and these are things that AI cannot yet touch.
04
people with excellent management skills
Peter Drucker said in the book "The Practice of Management": Enterprises should first consider the greatest strengths and weaknesses of human resources, and based on this, establish a work organization that is most suitable for the characteristics and limitations of human resources. Human resources have a characteristic that other resources do not have: the ability to coordinate, integrate, judge and imagine. Coincidentally, these qualities mentioned by the old man are exactly what AI does not have.
AI can write all kinds of neat copywriting, but it can't write a slogan that really makes people excited; it can write a scientific and rigorous work plan, but it can't make team members truly unite for this plan and overcome difficulties. These requirements are indeed a bit beyond the outline for AI, but they are the basic capabilities necessary for excellent managers.
Excellent managers usually have high emotional intelligence, strong empathy, precise judgment, and the comprehensive ability to observe and employ people. Can assess the situation well, organize personnel according to local conditions, integrate various resources, and complete various routine or sudden challenging tasks. They have both professionalism, leadership and personal charm, and they are all masters of insight into human nature, which is difficult for AI to achieve now and even in the future.
05
imaginative and creative person
ChatGPT may be able to give you a copy of the Starbucks coffee brewing process, but it cannot tell you what Starbucks should look like before Starbucks appeared. But, humans can. The basic outline of Starbucks actually comes from the "fantasy" of founder Howard Schultz.
After a trip to Italy, Howard couldn't forget the atmosphere of an Italian café. And on this basis, another richer picture is conceived. A coffee machine beeps, the air is filled with the aroma of coffee, and a space where people can sit around and talk happily. It was based on this imagination that Howard finally determined the basic style and structure of Starbucks.
Imagination is a magical ability unique to human beings. It can outline new and even unappeared scenes for human beings without the help of external forces. However, ChatGPT can only generate output based on existing data and algorithms, and cannot independently imagine and conceive new ideas. This lack of imagination and creativity, before AI can generate autonomous consciousness, it is still difficult to challenge humans.
The book "Deep Work" mentioned that unprecedented technological development and influence have brought about a huge restructuring of our economy. In this new economy, three groups of people will gain a special advantage: those who can use intelligent machines to do their jobs beautifully and creatively, those who are the best at what they do, and those who have the capital.
In Brother K’s view, we still have a chance to make a dash for becoming the first two types of people, even if we have “one of them” behind. The future has come, and we don't need to panic too much. Humans invented the car not to let us compete with the car to see who is faster, but to let us use the car to help ourselves go further. Although the car eliminated the coachman, didn't it also give birth to the driver? Just like any change, there may be various pains in the AI era, but it does not mean that the sky will collapse. There is a classic line in "The Wandering Earth 2", "I believe that the courage of human beings can span time, every history, present and future." I would like to encourage everyone.
-- END --
Send you a copy of "Top Institution ChatGPT Research Report" pdf
Follow the official account below and reply in the dialog box: gpt