On the other hand, low-skilled jobs are increasingly threatened by automation. Low-skilled jobs such as assembly line workers at manufacturing plants, delivery sorters, supermarket cashiers, and telephone customer service positions have come under extreme scrutiny for redundancy by automation and AI technology since their tasks are quite repetitive and standardized. For example, many blue-collar jobs in Amazon warehouses have been taken over by Kiva robots that operate sorting, handling, and packaging jobs of the work before increasing efficiency while cutting costs of labor. In the past five years, the number of robots in Amazon’s warehouses has jumped by over 50 percent while hiring for warehouse positions remains stagnant.

Low-skilled service jobs, however, are facing a compatible threat. For instance, banks and insurance companies have implemented AI robots to field customer queries and complaints on a very large scale; the hotel and restaurant industries are using automated checkout systems and smart food service ordering to cut down on labor needs. With the maturing of Natural Language Processing technology, customer service positions are expected to be automated more than 50 percent during the next five years.

In contrast, middle-skill jobs find themselves in an even bigger bind. Such jobs generally more commonly demand at least the basic skills and competencies, the experience but not certifiable capabilities; for example: administrative assistant, accountant, data entry, and technical support. Most of the tasks carried out by these roles are now achievable through automation and AI technology where a lot of their work can be done by algorithms or automated systems. For example, this has drastically reduced the number of middle-skilled roles in jobs like accountants due to the rise of RPA technology in the finance departments of enterprises performing activities like invoice review, reimbursement approval, and data entry.

A parallel dilemma lies in clerical jobs. Intelligent document-processing technologies such as Microsoft’s Cortana or IBM’s Watson are processing agenda meetings, scheduling, and email management automatically, massively diminishing the workload for administrative assistants. Often the job skills incorporated between high and low skills are unfitting for a swift metamorphosis into additively valuable work, hence leading to so-called “dumbbell” structures for employment demand: high demand for skills will reject the pseudo-skill jobs while low-skilled labor is on the rise.

Manufacturing is one of the domains that AI and automation technologies have wreaked terrible havoc on. Competition for basic assembly-line jobs is now intensely hiding from robots and automated machines. For example, Foxconn has installed more than 100000 industrial robots in its factories on the mainland of China to assemble and test mobile phones, computers, and other electronics. Such robots are far more efficient, and they can work non-stop throughout the day at a relatively low marginal cost as compared to human workers. The World Economic Forum predicts that by the year 2025, over 20 percent of manufacturing jobs will have been replaced globally by automated systems.

Aside from assembly jobs, quality inspection jobs, logistics jobs, and warehousing jobs are also under threat from AI. Visual-emotional identification systems and automated logistics machines can perform inspections, sorting, and inventory management quickly, which has greatly reduced labor costs. With the development of intelligence manufacturing (Industry 4.0), the need for manufacturing labor has changed from unskilled labor to highly skilled talents with equipment management and data analysis skills.

The financial sector is another sphere of AI technology that is immensely affected: in risk control, customer service, and asset management. In the past, financial risk control was primarily dependent on professional risk managers and human audit, however; risk control systems powered by machine learning can, within seconds, efficiently process enormous amounts of data and give precise assessments of risks for loans and investments. For instance, J.P. Morgan Chase developed a “program, for instance, within seconds can complete review work of thousands of pages of loan contracts with 360 times the efficiency of conventional manual work, finally achieving accuracy better.

Likewise, traditional jobs in customer service are being supplanted by AI. Intelligent customer service systems grounded in natural language processing (NLP) can answer inquiries at any hour of the day and consistently enhance service quality through learning. For example, Erica, the AI assistant at Bank of America, has dealt with over 100 million customer requests since its introduction across various business scenarios, including account inquiries, transaction history, and financial advice. This rising phenomenon would greatly lower the pool of nominally skilled jobs in the financial sector.

In asset management, the application of AI algorithms has made “quantitative trading” and “Robot-advisory” gradually mainstream. AI systems, by analyzing market data in real time, could offer better investment advice and risk assessments. This poses a serious challenge to traditional financial analysts and investment consultants.

Rapidly evolving AI technologies have not only altered employment architecture but also further underlined salary misalignment, bringing social inequality to the foreground. In this new wave of technology, highly skilled people have enjoyed splendid income rises, while for the low-skilled, income has remained stagnant or even fallen, owing to unequal distribution of technological benefits.

In the AI-driven economy, advanced skilled employees, especially technicians with special skills like data analysis, machine learning, and algorithm design, have turned into the “darlings” of the labor market enjoying not only generous salaries but also bonuses and extensive projected company stock options. According to the World Economic Forum, salaries for high-skilled jobs are projected to go up by more than 15 percent in the next five years. AI engineers in the San Francisco Bay Area, for instance, already command a base average annual salary of $ 150,000 level much higher than other traditional industries.

The technique of earnings for low-skilled employees has plateaued if not declined. All kinds of low-skilled jobs in traditional manufacturing, retail, and services are battling for survival in the low-skilled labor market, so the pay rate for the remaining has been suppressed. Predictions come to conclude that in five years, the pay for low-skilled jobs will drop by around 10 percent. Retail cashiers in the US, for instance, are not only facing job cuts with the introduction of automated checkout systems, but also their wages have stopped rising over the years.

The effects of every such trend have been to increase inequality in the wealthy piling up at the other end. It has become a parade for a couple of top-skilled individuals to reap all the technology rewards, while the pay for low-skilled workers has been brought down and the cost of living continues to rise, making the gap between them and the wealthier more apparent.

Apart from the difference in skill sets, other regional differences are also a strong determinant of wage disparity. In the developed regions, the agglomeration of high-tech establishments attracts large numbers of highly skilled personnel, forming a virtuous cycle of high income and consumption. High-level jobs in commercial cities such as San Francisco, New York, Singapore, and Shanghai, for instance, pay more than the national average. This not only boosts local economic growth but also further attracts capital and talent.

Therefore, in the backward areas, automation and industrial relocation have dealt traditional manufacturing jobs a fatal blow resulting being fewer jobs and stagnant wages. Long-established steel and automobile manufacturing industries in the Rust Belt of the United States have declined during the far-reaching phase of automation, with rising local unemployment and poverty rates. In China’s old industrial northeast and India’s traditional agricultural states, the plight continues to this day. Skilled laborers and high-paying jobs have been attracted to a few cities, while unskilled workers are stuck in regions where their wages are extremely low and opportunities for unemployment are few. Though these regional inequalities lead to an unjust distribution of inputs, they further heighten inequalities with respect to the distribution of other social resources. Regions with a high income would further invest in providing education, health facilities, and other public services which would further facilitate the fostering and accumulation of highly skilled talent and create a “Matthew effect” whereby it makes the strong stronger.

According to the report of the World Economic Forum, high-skilled occupations will have their salaries increase by more than 15 percent in five years, while low-skilled occupations have a potential risk of dropping off with the figure of 10 percent. This reinforces that in the wake of artificial intelligence-driven technology, the advantages of pay disparities and social inequalities are rapidly changing. Further, McKinsey has pointed out that the growing technology dividends are likely to generate conflict unless properly attended to and this may even hinder economic sustainability.

In the future, the growing salary gap will accelerate the agglomeration of highly skilled talents into big cities and metropolitan areas, making life more difficult for the poor regions and the less-skilled workers. Solving this problem of country re-employment training, regional economic development, and reforms to the social security system have turned into an important issue that must be faced by governments across the world.

With the rapid development of artificial intelligence technologies, low-skilled workers and traditional jobs are going through an unprecedented impact, thus making the role of governments crucial in the re-employment strategy. Proper strategies for effective intervention would include education and training, social security and subsidies, and the establishment of policies.

First, education and training are primordially one of the focuses means to address the effect of AI. With the growing need for high-skilled jobs, the government should invest in science, technology, engineering, and mathematics (STEM) education, particularly in primary and secondary schools introducing courses in programming, data analysis, and basic AI to educate many on AI applications. For example, some high schools in California, USA, have included computer programming as a compulsory course, thus significantly improving students’ adaptability to technical jobs. In addition, it will be necessary for the government to fund free training programs for low-skilled workers, focusing on data processing, operations of automated equipment, and basic programming. A prime example of such a case would be Singapore Skills Future. Funded by the government in partnership with corporate entities, Skills Future provides lifelong learning subsidies for training in AI. Statistics show that over 500,000 Japanese employees have undergone re-employment or job change training since the inception of the Skills Future initiative. This effectively forestalls low-skilled workers falling out of jobs due to automation. Secondly, improving social security and subsidy policies is particularly critical to balancing short-term unemployment. As automation improves, it becomes problematic that low-skilled workers in some industries will change their skills over a short time. Therefore, the government needs to build up a better social security system. The reformation of the current unemployment insurance system is a good approach. For example, one could use the unemployment insurance fund for occupational skills training, whereby the unemployed could be retrained during their unemployment via “reemployment training vouchers.” Germany has developed a successful experience in this direction — the German government provides benefit assistance to the unemployed for a maximum duration of two years, and in such duration, modified unemployment insurance is expanded to create a special extended aid for re-training in areas like AI, big data, and the operation of automated equipment. Such models have effectively mitigated the vaccine of job losses created by technological changes. Moreover, the government should extend re-employment bonuses that encourage firms to place unfired workers first. For instance, the Japanese government subsidizes wages with an upper limit of 50% of an employee’s salary. This significantly improves the burdens of hiring entailed by the enterprise, succeeding in stabilizing the job market. Finally, forward-looking policy design is key for ensuring the sustainability of re-employment strategies. The government can create a lifelong learning account system where an amount is deposited into learners daily to cater to vocational training and educational course dues. Finland is already piloting lifelong learning accounts in some regions, providing subsidies of around 1,000 euros a year for courses in data analytics, programming languages, and AI applications. Such a policy can ease the financial burden on the individual and promote skill renewal. In addition, governments should induce flexible reforms in the labor market, such as legislation to support flexible work and remote work, which increase job adaptability. The Netherlands’ flexible employment policy is an excellent account of that. By securing the legality and the benefits of part-time, and remote work, the Dutch unemployment rate has been kept low for some time now, thus having successfully deflected the unemployment pressure borne from technological progress.

In conclusion, education and training, social security, and policy support are the three main strategies through which the government will respond with respect to the impact of AI and further the policies of re-employment. Only the coordinating promotion of these individual aspects can facilitate the alleviation of the impact on employment as engendered by AI, to ensure inclusiveness in social development and sustainability of economic growth.

No doubt the government plays a critical role in addressing the impacts of AI on employment. It works to reduce unemployment risks through training and education, social security, and policy support. However, the government alone cannot provide a panacea to this problem. There is also an indispensable need for private-sector enterprises and individuals to respond actively. The enterprises, in view of being the agents of technological change, must bear more social responsibilities they can assist employees to adapt to technological change through retraining, internal job transfer, and cooperative innovation. A systematic mechanism in AI skills training can ensure the transition of employees from traditional jobs to high-value-adding ones. For instance, the IBM New Collar Jobs program has successfully transitioned a number of employees into new positions through short-term training and internships in areas such as data analytics and cloud computing. Certainly, CSR implementation here would help provide career transition guidance and re-employment opportunities to defend the basic rights and interests of employees. Collaboration with universities and vocational training institutions in establishing internship and skills-upgrading programs can create talent aligned with AIera’s needs. For example, Google has collaborated with many universities to launch joint training programs along the AI-concentrated and cloud computing tracks, increasing talents’ adaptability and competitiveness.

In the face of AI-induced changes, it is necessary for the labor force to take charge of enhancing their skills and diversify their income source to be capable of competing in the work environment. Skills such as programming and data analysis will not only ensure the competitiveness of the person but also allow him or her to contend against the risk of a job change arising out of AI. Enrollments in AI courses on online educational platforms like Coursera and Udemy have greatly expanded, underlining the importance of skill enhancement. Exploring side businesses and freelancing can help reduce the risks arising from a single income stream; while developing boundary-spanning skills and a sense of lifelong learning can improve the temporality of workers and avoid some risk for having techno-social change. The employment challenge arising from the use of artificial intelligence will not be resolved effectively unless the government, enterprises, and individuals jointly establish a comprehensive re-employment security system that will further promote the inclusive growth of society. The intensity of AI could enhance employee productivity and put together advancements in technology ethics and equity. Mechanisms of protection for the rights and interests of low-skilled workers during remedial processes of automation have conclusively become a serious process of social challenge requiring faster solution approaches. If the distribution of technological dividends favors high-skilled workers, for a long time running, the position of increased unemployment and salary declines of low-skilled workers is just a precursor for more changes in that direction. Apart from that, there are new inequalities arising from global competition and regional divergence. These developed countries use their vantage position in AI technology to minimize increased unemployment risks and salary reductions of low-skilled workers through the adoption of technology in other parts of the world. Under different geographical advantages, the odds are against new entrants in job markets. For example, countries in Africa and Southeast Asia are facing increased unemployment risks due to aid, manufacturing, or industries, all through automation. The future landscape will see a full merger of Artificial Intelligence and human resource management, creating the pathway for lifelong learning and flexible employment. Using intelligent talent management systems, companies have the advantage of making a more specified allocation of jobs and resources, however, there is still a need for politicking around it, coupled with provisions for training and continuous learning, to sustain fairness and balance in that job market. Balancing the use of technology, ethical values, and societal protection of minority groups in unleashing the general principles toward achieving inclusive and sustainable growth is going to be of central focus in the future thereby allowing productive culture-enabling transformation in the global valued systems in times of rapid socio-economic shifts.

In a nutshell, AI is changing the nature of labor, labor markets, and compensation systems while enhancing productivity. Among the skilled workers, some obvious technological windfall is felt, while low-skilled workers are battling the twin evil of joblessness and salary erosion. To address this challenge, governments, businesses, and individuals should come together and promote skills retraining and social protection reforms. Whereas the governments can wield their influence through the systems of education and social security, the enterprises can use their influence through retraining while generalizing social responsibility, and the individuals should upgrade their skills and plan their careers: these should come together to create a re-emplacement system commensurate with the age of AI. Perhaps only human and machine collaboration coupled with skills upgrading can guarantee inclusive growth in an era of artificial intelligence, ensuring continuity in technological development while maintaining social fairness and stability.

By Qingning Zhao