What AI development might look like in China’s socialist system

There has been much discussion of the US-China technological race, but little around how AI might evolve under China’s economic system
What would the model be in an AI-empowered world where the technology itself becomes a production resource?

Simon S.H. Chan

Published: 8:15am, 13 Jul 2023

Why you can trust SCMP

At the World AI Conference in Shanghai earlier this month, Tesla CEO Elon Musk said China is in a strong position when it comes to the development of artificial intelligence (AI). But what does this mean?

China has made great strides in building its AI capabilities and industry, and its companies and institutions made up seven of the 10 largest patent owners in AI and machine learning last year. According to official data, China has an edge in areas such as computer vision and natural language processing, with a core industry worth over US$71 billion, and more than 4,200 AI companies accounting for 16 per cent of the global total.

China’s AI sector has benefited significantly from the 14th five-year plan, spanning 2021-2025 and outlining the steps the country will take to build a modern socialist society. It places great emphasis on developing fast-growing technologies, with AI being a key area.

In its position paper on AI governance, the government said, “China is committed to building a community with a shared future for mankind in the domain of AI, advocating a people-centred approach and the principle of AI for good”.

There has been no shortage of discussion around the US-China technological race – but little around how AI might develop and evolve under China’s unique economic system.

China is deeply rooted in socialism, which permeates every aspect of life, values and policymaking. Globally, socialism as a political movement emerged in response to the economic and social upheaval of the industrial revolution and subsequent wealth gap. Interestingly, AI has the potential to be equally transformative and its impact on jobs is a concern for governments globally.

Under China’s socialist system, there is a social contract between the government and its people whereby the government is responsible for controlling and providing production resources, while individuals and businesses use these resources to earn a living, and in return pay taxes.

Throughout China’s economic transformation, the government has maintained control over the production resources, such as the critical infrastructure of energy, telecommunications, shipping and rail, through state-owned enterprises (SOEs). This ensured the government could generate a profit and reinvest in national development. Despite market reforms and the integration of capitalism, China’s social contract remains intact.

What would the model be in an AI-empowered world where the technology itself becomes a production resource?

There are many scenarios and possibilities. For instance, in the same way that SOEs are responsible for energy production and distribution, could they also be responsible for the quantum computing power and data centres that power AI? Or could the government create AI models for businesses and individuals to build value from?

04:12

How Huawei's use of 5G and AI is transforming China’s coal mining industry

Although there has been no specific mention or definition of what constitutes a production resource, the draft AI policy from the Beijing municipal government provides some hints as to how things might evolve. The first point of the blueprint calls for closer collaboration between cloud providers, the sources of computing power and research institutions. As part of this, it proposes a centralised platform, backed by the state, which would allocate public cloud resources to users on demand.

When it comes to AI research, rather than providing incentives or subsidies to the private sector as seen in Western countries, the Chinese government is focused on developing AI technologies through a network of regional, government-sponsored AI laboratories. This opens up the possibility that such labs could one day become SOEs, with the AI models they produce becoming national production resources.

Social stability is a fundamental pillar of Chinese society, and mass unemployment resulting from rapid AI adoption is not an option. China’s current policies mainly aim to promote AI development to maintain the economy’s competitiveness and to realise the benefits of enhanced economic productivity. Therefore, there needs to be a mechanism to achieve a healthy balance.

05:27

‘Socialism with Chinese characteristics’ explained

At the same time, China is focused on building economic self-sufficiency. President Xi Jinping has called for the establishment of a “strategic baseline” for key primary commodities including energy, grains and minerals.

As a result, we may see people encouraged to work in the primary and secondary production sectors. However, rather than a large number of young people becoming farmers and factory workers, it is more likely that they will be trained to build, operate and maintain robots and AI platforms to serve these sectors. The government is already promoting vocational education that teaches students how to operate AI-enabled robots – thus turning blue-collar jobs into “new collar” jobs.

For a country so proud of lifting over 800 million people out of destitution and eradicating absolute poverty in 2021, a return to the primary and secondary sectors may feel like a step back. Furthermore, it may inadvertently create a socioeconomic gap where wealthy students can afford to go to university, leaving others behind, which goes against the social change China is trying to generate.

How China’s jobless youth were raised to have unrealistic expectations

This begs the question of whether economic policies, such as a universal basic income, funded by a dividend extracted from increased productivity from AI, could be introduced to maintain people’s standard of living in the distant future.

Back in the 1990s, China experimented with a minimum rural living standard guarantee called “dibao”, which was expanded nationwide in 2007. But poor distribution, administrative deficiencies and a lack of reliable data meant the programme has not had a substantial impact, with the World Bank estimating that the programme only narrowed the poverty gap by 6.5 per cent.

However, with the government’s push for technologies, such as blockchain to drive greater accountability and transparency, better national digital infrastructure and leaps in AI, we may see the effectiveness of such programmes dramatically improve.

Simon S.H. Chan is the head of technology of the Greater Bay Area at Edelman. He is based in Hong Kong