Chinese Government Calls on Cloud Providers to Support AI Companies

Image credit: Photo credit: Bu Xiangdong/Qianlong.com/VCG / Getty Images (image has been modified)

China is ramping up state support to spur its own AI pioneers as Western tech’s large-scale language models have the potential to disrupt everything from marketing to education to coding.

The Chinese government is currently soliciting public opinion on a draft policy aimed at developing artificial general intelligence (AGI), a category of AI that can theoretically perform all human tasks. In a nutshell, the policy’s goal is to strengthen AI companies by getting more support from cloud providers and data companies.

It is not uncommon for capital cities to be at the forefront of policy-making in emerging industries. Beijing, for example, is the first to allow unmanned robo-taxis to transport passengers on open roads, subject to certain restrictions.

The AGI Blueprint presents an action plan in three main areas: compute power, training data, and applications.

The first strategy requires Close collaboration between cloud providers, the source of computing power, and universities and companies, which consume large amounts of processing power to train large-scale language models, multimodal learning, and other AI. This policy proposes a centralized state-sponsored platform that allocates public his cloud resources to users based on demand.

Alibaba took the top spot, accounting for more than a third of China’s cloud infrastructure services spending last year, according to market research firm Canalys. Huawei, Tencent and Baidu followed.

The second strategy acknowledges the lack of high-quality Chinese data and encourages “conforming cleansing” of such datasets. This includes anonymizing data, presumably in an effort to meet China’s strict new privacy laws. As we’ve seen how OpenAI relies on Kenyan workers to manually label training data and remove harmful text, this process will undoubtedly take time and effort. .

Beijing’s Big Data Exchange, launched in 2021 by the government to facilitate data trading across various aspects of society, will assist the process of data procurement.

Finally, the policy presents a list of potential pilot applications of AI, ranging from medical diagnostics, pharmaceutical manufacturing, financial risk management, transportation to the use of AI in urban management.

The proposed policy also touches on the importance of software and hardware infrastructure for AI training. The latter is striving to strengthen innovation in key technologies such as semiconductors amid intensifying competition between the United States and China.

The US has already restricted exports of Nvidia’s powerful H100 AI chip to China. In response, Nvidia came up with a less powerful processor for China to circumvent export restrictions. Domestic companies such as tech giant Huawei and startup Biren are also working on alternatives to Nvidia.