The impact of AI becomes a reality! The global employment landscape is accelerating reshaping

Microsoft's new round of adjustments for nearly 60,000 employees has intuitively put the contradiction between artificial intelligence (AI) and its impact on employment before the world. Since the beginning of this year, a large number of multinational companies, from large North American software manufacturers to European financial institutions to Asian e-commerce and manufacturing leaders, have used AI to replace more standardized positions. These companies have released a clear signal: AI is not just a tool to improve efficiency, its impact on employment has changed from long-term prediction to current reality, and the global employment pattern is accelerating in its reshaping.

Job changes at technology giants are often a bellwether for the labor market. In the past, industrial automation mostly affected simple manual labor positions. Now, under the influence of large AI models, agents and automation systems, the scope of replacement extends from assembly lines and basic customer service to document processing, financial analysis, auxiliary medical care, and junior programming. A large number of white-collar jobs have been involved, and the employment structure, income distribution and social governance of various countries have encountered new challenges. Analysis by the International Monetary Fund shows that about 60% of jobs in developed economies will be affected by AI to varying degrees. In emerging markets and low-income countries, the proportion of jobs affected by AI is expected to be 40% and 26% respectively. Another agency said that in the next 10 years, the reshaping of the U.S. job market by artificial intelligence will gradually enter a stage of deep adjustment, and a total of 15 million jobs may be replaced. The scope of substitution further extends to mid-range professional support positions. Standardized positions such as basic financial accounting, basic teaching and research assistance, routine personnel attendance accounting, technical function testing, and junior legal document compilation will face the pressure of automation.

This time AI replaces employment. The most significant change is that it breaks the traditional pattern of "blue-collar workers being affected first". Take the manufacturing industry as an example. In the early days, industrial robots mostly performed fixed and repetitive processes. Now with machine vision and intelligent scheduling, some production lines can adjust their own pace, identify faults, and optimize production schedules. A large number of positions with relatively fixed work contents have been consolidated and compressed.

The service industry has also been greatly affected. In industries such as finance, insurance, and cross-border e-commerce, intelligent systems have already taken over routine services such as basic customer service, document entry, and simple claims settlement. In the fields of knowledge and creativity, primary manuscript writing, basic translation, simple code writing, etc. are increasingly being completed by large AI models.

Research and analysis by some international institutions believe that middle-skilled jobs that were considered relatively safe in the past are now becoming the focus of AI replacement, giving rise to an obvious "skills polarization" phenomenon. At one end are positions that highly rely on complex decision-making, deep creativity and human empathy, which are difficult to replace by machines. On the other end, new professions that are good at controlling AI and capable of human-machine collaboration are rapidly expanding. Jobs caught in the middle and relying on standardized and repetitive knowledge continue to be squeezed.

National differences cannot be underestimated either. Although developed countries have strong investment in technology and sufficient retraining resources, they have a dense concentration of white-collar jobs and the impact often comes earlier. The labor force in some developing countries is mostly concentrated in low-skilled and low-value-added sectors. Without forward-looking preparations, they are more likely to face rising unemployment risks and a widening income gap.

In response to this accelerating employment substitution impact, it is obviously undesirable and impossible to try to slow down the pace of intelligentization. The key is to promote the education and training system to become lifelong, dynamic, and coordinated, and to make a series of institutional arrangements as soon as possible. For example, we should get rid of the traditional "once-education-for-life" model, incorporate artificial intelligence literacy into basic education, and allow learners to use intelligent tools as collaborative partners instead of just learning unchanging fixed processes. Germany continues to update vocational training content, bringing industrial intelligent operation and maintenance, robot debugging, and human-machine collaborative training to corporate sites. Singapore has launched a national skills improvement plan to provide ordinary workers with large subsidies or even free short-term courses to help clerks and grassroots technical workers master new and easy-to-use skills such as prompt engineering, intelligent quality inspection, and data verification. These practices and experiences are all valuable for reference.

As the main body of technology application, enterprises should also assume corresponding social responsibilities. Priority should be given to "enhanced" technology solutions, allowing AI to take over boring, dangerous, and highly repetitive tasks, and focusing on improving workers' creativity and new skills, rather than simply laying off people and reducing costs, pushing all transformation costs to workers.

At the same time, social security and risk buffering mechanisms must be improved. Jobs are changing faster and faster, and the cycle of workers' skills conversion is generally longer than the pace of enterprise adjustment. This requires continuous improvement of unemployment security, training subsidies, and transitional employment support to build a coherent channel from unemployment, retraining, to re-employment. For groups that find it difficult to complete the transformation through short-term training, special assistance must be provided to prevent technological changes from directly transforming into social divisions and conflicts.

At the global level, it is even more important to strengthen international dialogue and multilateral coordination. Developed countries can share their transformation experiences with developing countries and promote the open sharing of inclusive technological resources. Multilateral platforms such as the International Labor Organization and the World Economic Forum can jointly explore and formulate AI employment guidelines that take into account efficiency and human rights, prevent lowering labor standards in the name of technology, and avoid an unfair pattern in which "dividends are concentrated in a few countries and the costs are borne by vulnerable groups."

Looking further, AI is essentially not just a replacement tool, but also a new engine that drives productivity leaps. A large number of new positions are born precisely from human-machine collaboration. AI trainers continue to optimize the output of large AI models, industry experts rely on large models to conduct scientific research and demonstrate solutions, and factory technicians use intelligent systems to complete equipment diagnosis and process optimization... These new professions combine human experience, judgment, creativity, and empathy with the advantages of machines in high-speed computing and processing massive data. In the final analysis, the development of intelligence is not about who replaces the other, humans or machines, but who can reshape human capabilities faster, pushing workers to shift from simply performing tasks to controlling, verifying, and guiding intelligent systems.

The technological revolution will not stop, and the replacement of global jobs by artificial intelligence will continue to accelerate. For all countries, the best option is to proactively seize the window period, rely on education reform to open up skills transfer channels, guide enterprises to seek a balance between increasing efficiency and stabilizing jobs, use social security to support the bottom line of transformation, and rely on international cooperation to narrow the digital divide between the North and the South.

When more countries jointly explore this transformation path that is warm, inclusive, and collaborative, AI can break out of the single narrative of "replacing employment" and become an important force that generates higher-quality labor and promotes common global development. The dividends of this smart transformation can benefit every worker in the world.