As Shanghai’s annual political session began this week, a new proposal from one delegate called for more robust controls on social media platforms by using artificial intelligence. While the delegate, Chen Le (陈乐), chairman of a local business management and information systems firm​​​​​​​​​​​​​​​​, said the technology could address growing government concerns about misinformation and privacy violations, the moralistic and political overtones were unmistakable. Chen suggested that AI could help monitor “information leaks” (爆料) through public accounts (公号) and ensure a “healthy ecology” online. 

The supervisor (监事) and top shareholder of Wanrui Puying Group (万瑞普盈集团), Chen is a member of the Shanghai Municipal Committee of the Chinese People’s Political Consultative Conference (CPPCC), the country’s top political advisory body that brings together representatives from various political parties, organizations, and ethnic groups. 

In his proposal, “Suggestions on Introducing AI Technology to Strengthen Self-Media Content Review,” Chen outlines four main recommendations to regulate China’s self-media platforms (自媒体平台), which over the past decade have become crucial channels for information sharing and the expression of opinion in China.

First, he calls for AI technology implementation for automated identification and review of content posts and reposts. This system, the proposal said, would use natural language processing and image recognition to conduct real-time monitoring of “information exposures” (爆料) and other sensitive content. 

Second, the proposal recommends a classification and tracking system for published videos. This system would employ AI to tag content as “genuine exposure” (真实爆料), meaning that the content was credible, “false information” (虚假信息), or “privacy violation” (隐私泄露). According to Chen, this would enable better management and accountability. The system would embed mechanisms to trace content back to the original publishers and combat rumors and misinformation. 

Chen Le’s third recommendation focuses on leveraging existing automated review capabilities from established news media platforms. This would involve cooperation between state media organizations, industry associations, and universities to share technical resources and expertise — ultimately reducing manual review costs while improving accuracy.

The final recommendation in Chen’s proposal is where the less techy aspects finally come in. Mirroring frequent calls from internet control authorities, he suggests a renewed emphasis on industry self-regulation and strengthening legal frameworks. This would also involve stronger penalties for violations and enhancing public education about responsible social media use.

Along with 32 other delegates in Shanghai this year, Chen also joined a proposal that critiques Shanghai’s Lunar New Year cultural offerings as lacking distinctiveness compared to celebrations in other cities. AI also figures in that proposal, which argues that it could be used to “incorporate intangible cultural heritage.”