In 2028, the world will witness a pivotal moment in the AI arms race between the United States and China. The outcome of this competition will shape the future of global AI leadership and the rules that govern its development and deployment. The stakes are high, as the capabilities of AI systems will have profound implications for economic, military, and societal advancements. This article delves into two contrasting scenarios, highlighting the critical juncture we face and the actions required to ensure democratic nations maintain their lead in AI innovation and safety.
The AI Leadership Race
The competition for AI leadership is a complex interplay of technological advancements, policy decisions, and geopolitical strategies. At the heart of this race is the concept of 'compute,' the advanced computer chips that serve as the foundation for training and deploying AI models. The United States, with its innovative ecosystem of companies, holds a significant advantage in compute leadership, thanks to bipartisan support for export controls and a commitment to safeguarding American innovations.
However, China, under the jurisdiction of the Chinese Communist Party (CCP), is not far behind in terms of model intelligence. The CCP's AI labs have world-class talent and are exploiting loopholes in US export control policies, engaging in large-scale distillation attacks to replicate American innovations. This illicit practice allows them to stay close to the frontier, posing a significant challenge to US dominance.
The race for AI leadership is not just about intelligence but also about the ability to integrate and deploy AI systems across various sectors. China's 'AI+ Initiative' and its focus on 'embodied intelligence' demonstrate a strategic push to integrate AI into its economy and state apparatus, potentially gaining an edge over democracies in terms of global adoption and cost-effectiveness.
Two Scenarios for 2028
Scenario One: Democratic Leadership
In this optimistic scenario, the US and its allies successfully defend their compute advantage. Policymakers tighten export controls, disrupt China's distillation attacks, and accelerate AI adoption in democracies. This leads to a commanding lead in model intelligence, adoption, and global distribution. Democracies set the rules and norms for AI, ensuring safety and ethical considerations. Engaging with China on AI safety is a priority, and this scenario offers a promising path towards a collaborative and responsible AI future.
Scenario Two: Authoritarian Triumph
Alternatively, if the US fails to act, China's AI labs could catch up and even surpass American capabilities. Without tightened export controls and a lack of engagement on safety, China's AI ecosystem gains momentum. This scenario sees AI norms and rules shaped by authoritarian regimes, with models enabling automated repression and military advantages. The US, despite its initial compute advantage, loses its lead, and the world witnesses a disturbing shift in power dynamics.
The Imperatives of Staying Ahead
The competition for AI leadership is not merely a technological race but a battle for values and norms. Democracies must lead in AI development and deployment to shape the future of this transformative technology. A successful transition to AI-driven advancements in medicine, invention, and economic growth depends on where the most capable systems are built first. Responsible governance and safety measures should be at the forefront of this journey.
The CCP's aggressive pursuit of AI leadership, driven by state investment and a desire to establish China as a leading AI power, poses a significant threat. The CCP's use of AI for censorship, repression, and military enhancement highlights the risks associated with authoritarian control over AI.
Ensuring Democratic Leadership
To secure a favorable outcome in 2028, the following policy actions are crucial:
- Closing Loopholes: Tightening export controls and enforcing regulations to prevent illicit access to US compute and model access will hinder China's progress and sustain the democratic lead.
- Defending Innovations: Punishing distillation attacks and facilitating threat intelligence sharing among US labs will extend the democratic lead in model intelligence.
- Exporting American AI: Promoting the global adoption of trusted American AI technology will deny the CCP an edge in cost and adoption.
In conclusion, the AI leadership race in 2028 is a critical juncture that will determine the future of global innovation and safety. By embracing responsible policies and maintaining a strong compute advantage, democratic nations can secure their position as leaders in the AI revolution, ensuring a beneficial and safe transition for humanity.
