Chinese Internet Basic Corpus 4.0 Released with 120GB of Data for AI Development

On September 15, 2026, Chinese Internet Basic Corpus 4.0 was officially released at the AI Security Governance Sub-forum of the 2026 China Cybersecurity Week in Jinan. The dataset contains 120GB of data and is now available on the Chinese Internet Corpus Resource Platform. Users can visit the official website of the Cyberspace Security Association of China, access the platform link, and complete registration and verification to download or request the data.

According to the association, the release was carried out under the guidance of the Cyberspace Administration of China. The association worked with the National Internet Emergency Response Center, along with Baidu, Zhongke Wenge, Kaipuyun, TRS, iFlytek, Zhihu and other organizations, building on earlier releases of Corpus 1.0, 2.0 and 3.0. Through a data-sharing mechanism established by the association’s AI security governance committee, new high-quality trusted data was aggregated and processed with strict data cleaning measures to form the 4.0 release, providing credible data support for large-model training and AI development.

A representative of the association said the launch is an important step in the collaborative construction of high-quality Chinese language corpora and further enriches the domestic supply of such data. The association will continue to work with the National Internet Emergency Response Center and other partners to expand Chinese Internet basic corpus construction across more sectors and strengthen the data foundation for the AI industry.