When we wake up, DeepSeek starts to "lead" civil servants?

In 2025, AI will not only become a civil servant, but will also begin to “lead” civil servants.

With the popularity of DeepSeek, many places including Laibin in Guangxi, Zhengzhou in Henan, and Foshan in Guangdong recently planned to introduce DeepSeek to assist leading cadres in decision-making.

What’s going on? Can AI “lead” leaders?

According to the Laibin Daily, on February 18, the Standing Committee of the Laibin Municipal Party Committee of Guangxi held a meeting and proposed that leading cadres at all levels should embrace and learn new technologies with an active and proactive attitude, learn to master the use of large models such as DeepSeek, use artificial intelligence to assist in decision-making, analyze and solve problems, and use "human brain + artificial intelligence" to enhance personal capabilities and improve work quality and efficiency.

According to a report in Foshan Daily on February 18, Foshan City recently carried out local deployment of the DeepSeek artificial intelligence system, connected it to Guangdong Governance·Foshan City Brain, and launched the "Government Governance Zone" of the Guangdong Government Easy Work Station.

In addition, Zhengzhou, Henan Province stated that it will promote the study and application of artificial intelligence technologies represented by DeepSeek; CS-DeepSeek in Changsha, Hunan Province will be implemented in Changsha's urban intelligent security management field; and Linyi, Shandong Province will take the lead in firing the "first shot" of connecting the province's government affairs system to DeepSeek.

Yun Jie, researcher at the Institute of Political Science of the Chinese Academy of Social Sciences and director of the Center for Public Administration Research , said in an interview with China News Service's National Affairs Express that this phenomenon reflects the deepening of government departments and leading cadres' understanding of artificial intelligence and their emphasis on its functions and roles, and also reflects a new trend in China's digital government construction.

"To a certain extent, it marks that my country's government governance model has entered a new stage of technological integration, but it should be emphasized that this is only an initial stage of technological integration of the government governance model," said Yun Jie.

From AI civil servants’ report compilation and statistics to AI pediatricians’ medical data analysis, the combination of “AI + human brain” is meeting the needs of various industries and fields and is being used by them.

Today, the local governments’ promotion of the “AI + human brain” decision-making model also reflects the government’s urgent need to improve governance capabilities and respond to complex economic and social issues.

Chen Jianwei, professor at the National Institute of International Strategy at the University of International Business and Economics, said in an interview with China News Service's National Affairs Express that as the urbanization process reaches a high level and data grows explosively, traditional decision-making models are unable to efficiently process massive amounts of information, and AI technologies such as DeepSeek can provide data-driven insights to help implement precise policies.

In addition, some places emphasize AI-assisted decision-making to improve work quality and efficiency, which shows the realistic consideration of local governments in pursuing maximum efficiency under limited resources. Chen Jianwei pointed out that from a strategic perspective, this is also an active choice for local governments to adapt to the wave of digital economy and enhance regional competitiveness.

However, the application of artificial intelligence in government decision-making is still in its early stages, and its main application scenarios are the collection and analysis of decision-making information and the preliminary selection of decision-making plans. In Yun Jie's view, these applications are still not mature enough and need to develop in the direction of more professional government artificial intelligence in the future.

"What needs to be emphasized in this process is that when applying artificial intelligence, government departments should not only focus on the efficiency and scientific nature of government decision-making, but also pay attention to the security of decision-making," said Yun Jie.

Of course, security is by no means the only issue that needs to be considered when AI is introduced into government systems. In the process of government digital transformation, how can local governments avoid the formalistic problem of "technology for technology's sake"?

Yun Jie believes that only by starting from the government's own problem orientation and needs can we provide a clear direction and strong impetus for the combination of government management and AI technology. If technology is introduced just to follow the trend, then this development lacks motivation.

Specifically, we need to consider the problems existing in traditional government management. Which of these problems can be solved through traditional institutional reforms and mechanism optimization, and which problems cannot be solved by traditional reforms and optimizations? In this case, new means and approaches need to be provided through artificial intelligence technology.

Chen Jianwei believes that demand orientation also requires AI applications to focus on specific issues, such as improving cadres' AI capabilities through special counseling, solving pain points in decision-making rather than blindly pursuing new things. To avoid formalism, local governments should establish an evaluation mechanism to ensure that AI results are quantifiable and verifiable, rather than just staying at the display level.

At present, my country's government governance model is moving towards the initial stage of technological integration. Many places have organized cadres to learn DeepSeek, also to improve the level of government services and ensure that grassroots governance keeps pace with the times.

As AI enters the government system, Chen Jianwei believes that this process will also face three major challenges: technology adaptation, data governance, and skills shortage.

First, the technical adaptation problem is reflected in the fact that the existing government systems may not be compatible with AI tools and may require customized development to match local needs;

Second, data governance involves data silos and quality issues. If the basic data is incomplete and data between departments is isolated, AI analysis may be distorted, affecting decision-making results.

Third, the skills shortage is manifested in the general lack of AI literacy among cadres, which requires long-term training and cultural transformation to achieve