China to create and implement national standard for large language models in move to regulate AI, while using its power to transform industries

The China Electronic Standardisation Institute, under the Ministry of Industry and Information Technology, will enact a local standard for LLMs
Baidu, Huawei, 360 Security and Alibaba have been enlisted by the institute to lead a special task force that will draw up the new LLM standard

Artificial intelligence

Ben Jiangin BeijingandAnn Caoin Shanghai

Published: 8:15pm, 7 Jul, 2023

Why you can trust SCMP

China has set up a new government body that will be responsible for implementing a national standard for large language models (LLMs) – representing the technology used to train artificial intelligence (AI) chatbots like ChatGPT – as Beijing seeks to minimise potential disruption from this field, while harnessing its power to help transform traditional industries.

The China Electronic Standardisation Institute, which is under the Ministry of Industry and Information Technology (MIIT), is currently in the process of enacting a local standard for LLMs to support the growing number of fresh AI development initiatives now under way across the mainland, the agency announced on Friday at the World Artificial Intelligence Conference (WAIC) in Shanghai.

The institute, which did not provide details about this national standard, said it has enlisted internet search provider Baidu, telecommunications equipment maker Huawei Technologies, cybersecurity firm 360 Security Technology and e-commerce giant Alibaba Group Holding to lead a special task force that will draw up the new LLM standard. Alibaba owns the South China Morning Post.

LLMs are deep-learning AI algorithms that can recognise, summarise, translate, predict and generate content using very large data sets.

People visit the booth of Alibaba Group Holding at the opening of the World Artificial Intelligence Conference in Shanghai on July 6, 2023. The e-commerce giant is part of a special task force that will draw up a national standard for large language models. Photo: Reuters

China’s latest standardisation initiative reflects how local authorities have extolled AI’s potential to help drive economic growth and become a useful daily tool, while maintaining caution about its risks and asserting regulation of the technology.

The latest commentary on the subject published last month by the People’s Daily, the official newspaper of the Central Committee of China’s Communist Party, indicated that the government would intensify efforts to unleash AI’s potential.

It also listed the areas where China could benefit from AI, including daily office work, biopharmaceuticals, remote sensing and meteorology.

After Microsoft Corp-backed start-up OpenAI released ChatGPT in November, Chinese Big Tech firms have been rushing to develop challengers with Beijing pinning its hopes on AI development to bolster industrial productivity and fuel post-pandemic growth in the world’s second-largest economy.

Will AI be China’s ace in the hole to surpass the US and become the top economy?

At the WAIC event, which concludes this Saturday, the China Academy of Information and Communications Technology (CAICT), another institute under the MIIT, said it was also working to boost domestic LLM development and manage potential risks from the technology.

“[We will] promote systematic breakthroughs and original innovation in large [AI] models,” CAICT president Yu Xiaohui said in a presentation at the conference on Friday.

He said that it was important to accelerate development in the “verification and application of core areas such as algorithms and high-performance chips”, while strengthening research and coordination in “risk governance” to make large AI models “reliable tools for society’s development”.

Tencent Holdings, Baidu, Huawei and Alibaba, have rolled out their respective LLM applications for wider adoption in various businesses. Tencent, for example, last month launched an LLM service targeting traditional sectors from finance to media.

Attendees watch a display on generative artificial intelligence technology at the Tencent Holdings booth during the opening of the World Artificial Intelligence Conference in Shanghai on July 6, 2023. Photo: Bloomberg

Still, internet regulator the Cyberspace Administration of China (CAC) has yet to issue a licence for any generative AI product in the country, even as Big Tech firms like Baidu, Alibaba and iFlytek have rolled out ChatGPT-like services on a trial basis.

Generative AI describes algorithms that can be used to create new content, including audio, code, images, text, simulations and videos. Recent breakthroughs in the field have the potential to drastically change the way people approach content creation.

All generative AI algorithms and products must go through security testing and review by the CAC before these can be made available to the public.

CAC director Zhuang Rongwen last month said at an event in eastern Shandong province that it was the government’s aim to “make sure AI is reliable and controllable”.

Post