For global AI regulation to succeed, China must have a seat at the table

The UK’s decision to invite China to the AI Safety Summit despite objections, and the resulting Bletchley Declaration, is a clear sign of progress
Global issues of this kind simply cannot be solved unless there is buy-in from China

Connor Horsfall

Published: 5:30am, 24 Nov 2023

Why you can trust SCMP

Chinese President Xi Jinping has just concluded his first visit to the US in six years. Yet despite positive readouts from the visit, relations between the West and China remain cold. Discussions of a G2 relationship feel like a distant memory and there remains a consensus across the West on “de-risking” supply chains away from China in areas critical to national security.

In the United Kingdom, accusations of espionage in parliament followed by an MI5 warning that more than 20,000 people have been approached by suspected Chinese spies have done little to rebuild trust. These tensions did not, however, stop the UK government from inviting China to the AI Safety Summit.

Chinese officials attended both days at Bletchley Park but were not invited to a public meeting on the safety and security risks from AI. Wu Zhaohui, China’s Vice-Minister of Science and Technology, called for “global cooperation to share AI knowledge and make AI technologies available to the public on open source terms”.

Many were concerned about China’s summit involvement. Former leader Liz Truss, in a letter to Prime Minister Rishi Sunak, argued that “no reasonable person expects China to abide by anything agreed to at this kind of summit”. This may well be true, but the UK was right to invite China and should be commended for trying to build consensus on AI risks.

China is second only to the US in AI investment and development, pouring in tens of billions of dollars annually, and has set 2030 as its deadline to become a global AI leader.

When it comes to regulation, China has moved more quickly than the United States, EU and UK. In July, China published new rules for domestic AI development for synthetically generated images and chatbots, following the issuance of regulations on recommendation algorithms.

The West has since caught up. US President Joe Biden recently signed an executive order featuring eight guiding principles on the safe, secure and trustworthy development of AI. The European Union is in the process of passing an AI act and the UK is developing its pro-innovation approach to AI regulation.

Without global consensus, however, attempts to mitigate the extreme risks associated with AI will be futile.

AI is increasingly used in autonomous weapon systems. This could have significant implications for how conflicts are fought and experts have expressed concern that these systems could make lethal decisions independent of humans. Their development needs to be carefully monitored and regulated.

This is why China’s AI development matters. The People’s Liberation Army has invested significantly in AI applications, including in robotics and swarming, and the trajectory of AI will in many ways be influenced by Beijing’s decisions. We need to fully understand how China’s AI ecosystem is developing, and the associated benefits and risks.

01:26

Israel installs AI-powered gun at West Bank checkpoint as riot control

The West must not fall into the trap of assuming China’s AI governance structures are irrelevant to the development of Western governance models. There will be lessons to learn from China’s regulatory developments, especially the feasibility of adopting particular regulatory approaches.

We also need to keep China at the negotiating table. Global issues of this kind can’t be solved without key actors in the room and China will continue to be at the epicentre of them all. Preparing for pandemics and tackling the climate crisis simply cannot be achieved without Beijing’s political buy-in.

China has, over the last decade, taken a more active role in global governance, building its international presence through the Belt and Road Initiative, where nearly US$1 trillion has been lent by Chinese financial institutions, and playing a greater role in conflict mediation in the Middle East.

China’s 12-point peace position paper on Ukraine and its brokering of the re-establishment of Saudi Arabia-Iran relations shows a willingness to play a more active role on the world stage, and Beijing may still play a pivotal role in the Ukraine and Gaza conflicts. This is a significant change from 2012, when China was mediating just three global conflicts.

The hope is that Beijing is now more willing to engage in discussions on new AI governance frameworks. The agreement on the opportunities, risks and need for international action on frontier AI at the Bletchley Park summit is clear progress.

South Korea and France are to host the next two AI Safety Summits in 2024.

We need international consensus on issues that affect all of humanity. China’s attendance at these summits would be a positive step for global prosperity.

Connor Horsfall is a China specialist and consultant at Shearwater Global