The summer of 2026 was expected to be a stressful job-hunting season for Li Jiawei, a university graduate from Ezhou City, Hubei Province. Yet, to her surprise, her path to employment was remarkably smooth. "After graduating, I kept searching for jobs. Thankfully, the human resources and social security staff registered my needs and called or messaged me whenever they found suitable positions," she said. In Ezhou, stories like Li Jiawei's are becoming increasingly common, thanks to a deep-seated transformation in employment service models.
Behind this seemingly simple "precise matchmaking" lies a profound shift, driven by the implementation of the "15th Five-Year Plan" for Human Resources and Social Security Development and the "Opinions on Accelerating the Application of 'AI + Human Resources and Social Security'." Traditional employment services are being redefined as cloud-based big data and grassroots "iron feet" (a term for dedicated on-the-ground personnel) work together, forging a new path that combines precision with a human touch.
AI Opens New Horizons for Employment Services
In Yushui District, Xinyu City, Jiangxi Province, a dynamically updated "human resources map" has become a strategic tool for grassroots employment services. This map clearly marks the skills and job preferences of unemployed residents in the area. "Previously, job hunting felt like looking for a needle in a haystack. Now, the platform recommends matching positions based on my profile," said Mr. Zhang, a former jobseeker. Guided by this map and with the help of community grid workers, he found a stable job as a quality inspector at a local electronics factory. This is a vivid example of how AI is empowering employment services across the country.
Under China's "AI + Human Resources and Social Security" strategy, several mature digital application scenarios have been developed and are now replicable. Intelligent job-matching systems break the traditional one-way information release model, analyzing job requirements and worker skills using occupational knowledge graphs to help "jobs find people." Precision profiling systems combine job-seeking history, skill levels, and training records to provide crucial data for employment assistance. Smart policy recommendation algorithms automatically identify users and push relevant policies on social insurance subsidies, entrepreneurship support, and skills training to targeted groups like the unemployed, new graduates, and migrant workers. For vulnerable groups such as those lifted out of poverty, zero-employment families, and older jobseekers, dynamic monitoring systems track employment status and provide early warnings.
In Xuzhou (Jiangsu), Quanzhou (Fujian), and Qingyang (Gansu), digital platforms serve as "digital hubs" for employment services. Quanzhou has built the "Quanzhou Employment" smart platform, breaking down data silos across departments to gather information on labor demand and supply. Qingyang has launched the "Smart Employment Qingyang" platform, covering hundreds of thousands of rural migrant workers, with online livestream job fairs providing over 180,000 positions annually. In the market-based human resources sector, AI is also transforming recruitment. Leading firms like China International Intellectech Group and Zhaopin have integrated smart interviews and digital human job matching into their processes. Using multi-agent systems, they offer 24/7 recruitment services, effectively addressing mismatches caused by poor communication skills in traditional interviews.
"Promoting AI-empowered employment policies and services during the '15th Five-Year Plan' period is essentially using digital technology to reshape the supply of employment services," said Cao Jia, a researcher at the Institute for Employment Research at the Chinese Academy of Labor and Social Security. "This is a necessary path for modernizing the employment service system, a key lever for improving the efficiency of matching human resources supply and demand, and a foundational step for implementing the employment-first strategy and supporting high-quality development."
Cao Jia noted that this is reflected in five key areas: first, responding to technological change to promote high-quality and full employment; second, strengthening the data foundation for employment services to improve decision-making; third, improving matching efficiency between supply and demand to alleviate structural contradictions; fourth, reshaping grassroots governance by combining technology with on-the-ground reach to reduce burdens and increase efficiency; and fifth, leveraging long-term employment data to serve high-quality development and national security.
Grassroots Outreach Brings Human Warmth
While the advantages of technology are clear, grassroots workers quickly identified a practical gap: algorithms can filter resumes but struggle to understand the personal difficulties behind a jobseeker's profile. Data can mark someone as unemployed, but it cannot capture the true willingness of "silent groups" to find work. Many older workers, people with disabilities, and rural residents lack the skills to use online platforms, leaving them excluded from digital services.
"Online platforms are convenient, but some people can't cross the digital threshold. Relying solely on online systems often means services are visible but unreachable," said Zhao Pengfang, a community employment worker in Taiyuan, Shanxi Province, highlighting a common challenge.
"AI is good at collecting labor market information, helping to shift employment services from passive response to proactive identification and precise supply. However, employment services are not just about matching information; the service recipients are not just data points," said Wu Maohua, a lecturer at Nanjing University of Finance and Economics.
Wu Maohua explained that the constraints faced by the unemployed are often complex and hidden, including objective factors like family care responsibilities and skill gaps, as well as waning confidence after repeated failures. While big data can show someone is "unemployed," it may not explain *why*. Furthermore, AI's judgment depends on data quality. An algorithm-identified "job match" does not guarantee the person's willingness to work or the employer's genuine need, nor does it ensure long-term job stability.
More importantly, much of the core work in employment services relies on face-to-face trust and communication. Qin Ying, an employment assistance worker in Shanghai's Chongming District, discovered this firsthand. An older long-term unemployed person in her area had developed severe self-doubt after repeated job rejections. Dozens of job postings Qin Ying sent online went ignored. It was only through persistent home visits and patient emotional support that she made progress. "The system only sees his unemployment and age, but not his inner feeling of failure. Without opening that emotional block, even the best job offers are useless," Qin Ying said.
These are the services technology cannot replace: going door-to-door and into businesses to understand real needs, providing psychological counseling, building trust, offering targeted support to zero-employment families and people with severe disabilities, and helping resolve labor disputes and practical problems. In some areas, an over-reliance on digital platforms has weakened grassroots outreach. Some local units mistakenly believe that entering data into a system constitutes completed service, only sending online messages to the unemployed without follow-up visits. Many jobseekers remain in a "passive information receiver" state, their needs unheard, and the structural imbalances between supply and demand cannot be resolved by technology alone.
Wu Maohua emphasized that home visits and face-to-face communication are essential for verifying a person's actual situation and job preferences. Job verification, rights reminders, and follow-up check-ins help ensure job quality, prevent risks, and resolve specific issues during employment.
Building a New Model of Human-Machine Collaboration
The "15th Five-Year Plan" for implementing the employment-first strategy calls for adapting to AI development to promote employment and entrepreneurship. Experts suggest exploring a collaborative path where "big data + iron feet" work together, with a clear division of labor between AI and grassroots services, creating an innovative model that goes beyond purely manual or purely online methods.
Cao Jia suggests that big data should be responsible for identifying "who to find, where to find them, and what to offer," while the "iron feet" should be the trusted, approachable personnel who can get things done. The machine should step back to act as a "lens," while people move forward to be the "fulcrum," enhancing both the precision and the human touch of employment services.
Specifically, this involves three key steps. First, break down data silos and build a strong data foundation, while also making grassroots service tools more user-friendly, such as developing mini-programs or self-service terminals that allow on-the-ground workers to carry data and tools with them. Second, build a full-chain closed-loop service mechanism. This includes data screening, household verification, policy matching, assistance implementation, and effect tracking, directly converting the "demand list" from AI algorithms into tasks for "iron feet" workers, ensuring seamless online-to-offline service. Third, ensure data security and preserve human judgment. Data use should adhere to the "minimum necessary" and "de-identification" principles, prohibiting algorithmic discrimination based on age, gender, or disability. A human veto power should be retained, allowing "iron feet" workers to override machine decisions with documented reasons. For vulnerable groups unfamiliar with AI, such as older workers and those with low skills, face-to-face service from "iron feet" personnel remains essential.
"The key is to form a closed loop: data discovery, grassroots verification, categorized service, and result feedback," Wu Maohua explained. AI should handle information collection, monitoring labor demand, screening targets, and performing standardized tasks like initial policy matching and job matching. Grassroots workers should focus on verifying information, understanding the jobseeker's willingness and real difficulties, and providing career guidance, training coordination, employer liaison, and support. To achieve this, data sharing and service feedback must be strengthened, online identification and offline services must be effectively linked, and necessary human service channels must be maintained. AI improves the efficiency of problem identification and resource allocation, while grassroots personnel provide professional judgment and personalized services, making employment services both precise, efficient, fair, and warm.
Several regions are already exploring this collaborative model. Quanzhou has established a closed-loop mechanism for data retrieval, cleaning, and synchronization, enabling near-instant data sharing between provincial and city platforms. This allows for co-management of platforms and authorized data sharing, while leveraging the "iron feet" of grassroots employment service officers with portable tools to extend data collection to the "last mile." In Huaiyuan District, Bengbu City, Anhui Province, the Human Resources and Social Security Bureau uses "home-based" employment service stations as a base, employing a "three-tier classification method" for employment supply-demand surveys. They create personalized "one person, one policy" files, build a precise employment big data database, and use a 3-kilometer integrated information platform for online data analysis, scientific service evaluation, and precise job guidance.
In the Xing'an League of Inner Mongolia, the "big data + iron feet" model is used to build both a database of jobseeker needs and a database of employer needs. The local gig market is integrated into the "Mongolian Quick Service" app, forming a "1+6+N" gig service model (1 league-level market, 6 county-level markets, and N township-level service points). Additionally, the "Xing'an Employment" Douyin livestream brand has been created, making job selection as easy as a touch on the screen.
Technology is moving forward, but the focus remains on the ground. AI gives employment services a digital boost, vastly expanding their reach and efficiency. At the same time, the "iron feet" of dedicated workers, who go deep into communities and connect with people, preserve the warmth and foundation of these public services. Continuously improving the "big data + iron feet" human-machine collaboration model is the only way to bridge the gap between supply and demand, ensuring that every worker can fairly share in the benefits of employment service resources.
Comments