Artificial intelligence (AI) is helping transform China's social security administration, making services more accessible, improving efficiency and strengthening risk management, according to a new report by the International Labour Organization (ILO).
Prepared in collaboration with China's Ministry of Human Resources and Social Security (MOHRSS), the report, From digital to intelligent social security administration, documents how AI is being integrated into social security services and identifies lessons for its responsible adoption.
Examining 45 cases across 15 provincial-level regions in China, the research finds that AI applications are concentrated in four areas: improving service delivery, supporting administrative operations, strengthening risk management and enabling better-informed decision-making.
Among the cases examined, 47 per cent focus on improving service delivery, while 33 per cent support administrative operations. Applications involving risk management and decision-making account for the remaining 20 per cent.
AI-powered chatbots, virtual assistants and self-service platforms are enabling round-the-clock access to information and services, helping individuals navigate complex social security procedures and understand their entitlements.
The report also documents significant efficiency gains. In Beijing's Haidian District, automated systems helped process more than 70,000 social security transactions in 2024, improving processing efficiency by over 80 per cent compared with manual operations.
In Guangdong Province, AI-assisted review systems had processed more than seven million cases across 34 social insurance services by July 2025, reducing manual workloads by over 640,000 staff hours.
Artificial intelligence offers important opportunities to improve social security services and strengthen institutional capacity, but technological innovation is not an end in itself. How these technologies are introduced and governed is critical to ensuring that they contribute to more effective, inclusive and people-centred services, while safeguarding the rights and livelihoods of those who depend on them.
Mikhail Pouchkin, Director of the ILO Country Office for China and Mongolia
Despite these advances, the report identifies important challenges, including the complexity of maintaining accurate AI knowledge systems, protecting sensitive personal information, addressing algorithmic bias and ensuring accountability.
It emphasizes that AI should assist rather than replace human decision-making, particularly in sensitive areas such as benefit eligibility and work injury assessments. It also stresses that AI-enabled services should complement, rather than replace, human and offline services to ensure that older persons, persons with disabilities and others facing digital barriers are not excluded.
The report recommends strengthening data governance, investing in workforce skills, developing appropriate safeguards and encouraging the sharing of successful practices across regions.
From digital to intelligent social security administration is the first publication in the ILO's new AI and Labour Governance in China series, which examines emerging applications of AI across labour-related public services and their implications for governance and decent work.