Introduction
Welcome to our show. Today, we're delving into the hot topic of AI and its relationship with our future. There has been a great deal of discussion recently, with various opinions. We've gathered information from big AI industry leaders, including warnings, research, analysis, and market perspectives. It seems that AI is particularly effective in work, especially entry-level positions, and its impact may be faster than we thought. Many predictions have a short timeline, with potential changes within a few years. Our task is to sort through this information, understand the concerns of those at the forefront, and consider what it means for each of us.
Before we begin, we'd like to introduce our audio program, which is part of WOW's YouTube video channel. If you're interested, you can search for "atwow.insight" on YouTube, subscribe, and enjoy the more comprehensive and visual content.
The Warning from Dario Amodei, CEO of Anthropic
One of the most eye-catching warnings comes from Dario Amodei, CEO of Anthropic. He has been working on AI for 10 years and has witnessed its rapid progress. Two years ago, it was at the level of a smart high school student, and now it's at the level of a smart college student and beyond.
Amodei sees both positive and negative aspects of AI. On one hand, he believes it can cure important diseases like cancer and Alzheimer's, provide cheaper energy, and bring many other benefits. On the other hand, he's worried about the impact on entry-level jobs in finance, consulting, tech, and other areas. He predicts that these jobs will be first augmented and then replaced by AI systems within one to five years, potentially leading to a serious employment crisis with an unemployment rate of 10% to 20%.
Amodei's warning is significant because of his position as the CEO of a top AI company. He's not an external commentator but an insider who has likely seen trends and internal data that others haven't. His prediction is also based on a specific logic, that AI will first enhance these positions to improve efficiency but then replace them.
The Perspective of Anthropic Employees
Amodei is not the only one at Anthropic with concerns. Employees within the company believe that even if AI technology were to stagnate, existing models could automate all white-collar work within five years. Their reasoning is based on economic efficiency, arguing that even if it requires manual training for each task, the long-term cost savings of using AI are significant.
This perspective is even more shocking because it's based on current technology, not some future super-intelligent AI. It makes the prediction feel more immediate and relevant.
The Article by Avital Bowit, Chief Editor of Anthropic
Avital Bowit, the chief editor of Anthropic, also expressed similar concerns in an article. He predicted that in the next five years, it may be his last working year, not because of personal reasons but because of the potential end of the current employment model due to technological change.
Bowit analyzed that AI's ability to replace is too fast, and every update makes it stronger and more common. He expects AI to eventually be able to complete all tasks with economic value, which could have a profound impact on the job market.
The Impact on the Labor Market
There is some data and online attention to support these predictions. For example, some companies are starting to skip recruiting entry-level employees and instead use AI to handle basic work. The employment conditions of recent college students are deteriorating, with an unemployment rate of 5.8% and a high level of underemployment. Even top business school MBA graduates are facing difficulties finding jobs.
Another unusual phenomenon is the increase in applicants for law schools, which analysts believe is a sign of people seeking more stable professions in the face of potential economic uncertainty.
The Experience Gap
The replacement of entry-level jobs by AI also leads to a deeper problem known as the experience gap. The traditional career development path involves starting at an entry-level position, learning skills, and accumulating experience to grow into an expert or manager. But if these entry-level jobs are replaced, how will young people or those new to an industry gain the necessary practical experience? This could become a long-term structural problem.
The Response from Industry Leaders and Policy Makers
Even former US President Obama has been concerned about the impact of AI on employment. He shared articles discussing the issue and called for public and sincere discussions about the benefits of AI, as well as the need to consider response plans such as universal basic income (UBI).
Some economists have also pointed out that the work AI is particularly good at is the same as what many young college graduates do when they first enter the company. This further highlights the potential impact on the job market.
The Corporate Perspective
From the corporate point of view, the internal strategies of companies are also being affected by AI. For example, the CEO of Fiverr sent an internal email to employees warning that AI is coming towards them and that every position in the company is at risk. He listed programmers, designers, lawyers, salespeople, etc., and proposed that AI will make simple tasks disappear, difficult tasks become simple, and impossible tasks become difficult.
Fiverr's CEO also set a time limit of a few months for employees to learn how to use AI tools, dig deeper in their professional fields, and become top experts. His email is a vivid representation of the AI priority business strategy within the company.
The example of Fiverr is not isolated. Other companies, such as Business Insider, have also announced layoffs, with a significant proportion affecting all departments. This suggests that the AI priority may be becoming a broader trend.
The Impact on the Physical Economy
Even in the physical economy, such as Amazon's warehouses, automation and AI are rapidly advancing. Amazon has been using robots for a long time, and now they are getting more and more powerful. For example, the Cardinal system can pick up and move items like a mechanical arm, the Proteus is a self-moving robot for transportation, and the Vulkan system is an automated storage and selection system that can handle individual products.
Amazon's goal is to eliminate fragmented, monotonous, and repetitive work in the warehouse, although there are still tasks that require human operation. This shows that the trend of automation and AI is not limited to white-collar work but also affects the physical economy.
The Views of Other Technology Leaders
In addition to Amodei and Kaufman, there are many other technology leaders who have expressed similar concerns or opinions. For example, Mustafa Suleiman, co-founder of DeepMind, believes that in the short term, AI may be more of a tool to strengthen human ability and improve productivity, but in the long term, it could lead to great social problems if the power of the market is completely unleashed.
Sam Altman, CEO of OpenAI, is relatively optimistic in public, believing in human adaptability and the creation of new jobs. However, he also revealed in a private conversation that the development of AI technology could eventually eliminate up to 70% of existing jobs. At a hearing in the US Congress, Altman emphasized the key difference between the technological revolution caused by AI and previous revolutions is speed, and proposed a response strategy called relay deployment.
Aravind Srinivas, CEO of Perplexed AI, pointed out that AI progress may cause a large amount of labor loss in the short term, and a company with a high value may no longer need to hire as many employees as in the past. He also raised the question of where the next generation of college graduates will find jobs and how people can improve their skills to adapt to this change.
Imad Mustaq, former CEO of Stability AI, used the metaphor of a pile of sand collapsing to describe the potential unemployment phenomenon caused by AI. He believes that unemployment may not happen gradually but may suddenly explode on a large scale.
Another former employee of OpenAI mentioned the fundamental differences that general AI (AGI) or super-intelligence may bring. Once AGI or super-intelligence appears, the situation will be completely different because of its strong learning and adaptability, which could potentially make it difficult for humans to find work that only they can do.
Tristan Harris, co-founder of the Center for Humane Technology, warned that the current state of AI development among people is likely to lead to large-scale economic instability and social unrest. He believes that many countries are accelerating deployment in the field of AI without fully understanding the risks, and this competition logic is wrong.
The World Economic Forum Report
The World Economic Forum publishes a report on future employment every year. The latest report found that the vast majority of companies expect AI and big data technology to greatly promote their business transformation in the next few years. It also predicts that a large number of administrative jobs will be lost, such as secretary, accountant, cashier, and telephone salesman, while new jobs will be created in areas such as AI and machine learning experts, sustainable development experts, green technology experts, commercial intelligence analysts, information security analysts, education, agriculture, and digital related areas.
The report emphasizes that the growth of new jobs is mainly concentrated in a few areas with high technical content or specific professional knowledge, and there is a significant skills gap. More than 60% of employees feel that their skills can't keep up, so large-scale skill improvement and re-training will become extremely important in the next few years.
The report also provides a detailed list of the fastest-growing and fastest-declining jobs, which is very helpful for people to do professional planning. The fastest-growing jobs are mainly related to digital technology, AI, data, and security, while the fastest-declining jobs are more traditional, process-oriented, and easily replaced by digital technology.
In addition, the report lists a core skill list that everyone should focus on learning and mastering by 2025, including analytical thinking and innovation, active learning and learning strategies, complex problem solving, critical thinking and analysis, creativity, originality, and initiative, as well as technical skills such as technology use, monitoring and control, technology design and programming, and AI and big data application ability.
The Complexity of Reality and Personal Response Strategy
While the predictions and warnings about AI's impact on the future are concerning, the process of complete automation may be more complex and slower than some of the most extreme predictions. There are several reasons for this, including the limitation of technology itself, the challenge of actual deployment, and economic factors.
In many cases, AI is more likely to play the role of strengthening humans rather than completely replacing them. Human-machine cooperation may be a more common model in the future. However, we still need to pay special attention to those areas with higher risks.
From the point of view of Amodei and others, as well as the World Economic Forum report, the areas with relatively high risk include entry-level white-collar positions in technology, finance, law, and consultation, especially those involving a large amount of data processing, information collection, document review, basic analysis, answering common questions, and carrying out standardized processes.
The areas that are relatively safe in the short term include jobs that require high-level physical flexibility and eye coordination, such as skilled vets, electricians, and hair stylists; jobs that require in-depth interpersonal interaction, empathy, and trust, such as psychological therapists, teachers, and high-level management; and jobs that require highly professional knowledge, complex judgment, and creativity, such as top scientists, artists, strategic consultants, and experienced doctors.
For ordinary people, the key is to embrace change, learn from AI, and improve their own unique value. The personal response strategy includes learning how to use AI tools effectively, paying more attention to the advantages of human uniqueness and improving related abilities, and being dedicated in the chosen field to become a top expert.
A Deeper Level of Thinking
Before we end, we'd like to leave a question for the audience to think about. If AI is really a fundamental labor replacement tool, and even relatively optimistic people like Sam Altman propose a new economic system and discuss the possibility of a UBI program, then just relying on our own efforts to adapt and improve may not be enough to deal with this structural and systematic change in the long term.
Our social structure, economic system, and even our definition and value perception of work itself may need to undergo more fundamental reflection and adjustment. This is a complex and important question that requires further exploration and discussion.
Thank you for listening to our in-depth discussion today. We hope this information has been useful and thought-provoking.
{# 
{# 
{# 
{# 
{#