VIDEO
HIGHLIGHT
Log InSign up
The U.S. and China Are Not In An A.I. Race | Interesting Times with Ross Douthat
SummaryTranscriptChat

The U.S. and China Are Not In An A.I. Race | Interesting Times with Ross Douthat

The AI Race: US vs. China

Overview of the AI Landscape

  • The discussion begins with a focus on the two leading countries in AI development: the United States and China, likening their competition to a cold war atmosphere.
  • The speaker asserts that currently, China is ahead in this race, emphasizing that it’s a binary competition between these two nations.

Differences in Approaches to AI

  • The U.S. focuses on developing Artificial General Intelligence (AGI), aiming for superintelligence that can outperform humans across various tasks.
  • In contrast, China's strategy involves multiple races: improving AI models while also prioritizing efficiency and accessibility through open-source initiatives.
  • Chinese companies are heavily investing in practical applications of AI, particularly robotics, rather than solely pursuing AGI.

Everyday Life with AI in China

Practical Applications of AI

  • In major Chinese cities like Shanghai and Beijing, autonomous delivery robots and waiter robots are beginning to change daily life experiences.
  • While not yet widespread, these technologies represent a significant shift compared to American interactions with physical reality.

Government Regulation vs. Corporate Innovation

Regulatory Environment

  • Unlike the U.S., where tech companies operate with minimal regulation, China's Communist Party exerts strict control over its tech industry.
  • Regulations include mandatory registration of AI models and adherence to censorship rules set by the government.

Hybrid Model of Innovation

  • Despite strict regulations, Chinese tech firms maintain some level of innovation and competition within the framework established by the state.

Key Players in China's AI Sector

Major Companies

  • DeepSeek is identified as one of China's closest equivalents to Western firms like OpenAI or Anthropic; however, many successful startups exist within constraints imposed by government policies.

Constraints on Development

Challenges Faced by Chinese Companies

  • U.S. export controls limit access to advanced semiconductors for Chinese companies, forcing them to rely on less capable domestic alternatives.

Efforts Toward Self-Sufficiency

  • Companies like Huawei are attempting to develop their own chip manufacturing capabilities but face challenges competing with Nvidia's technology.

Energy Resources as an Advantage

Energy Infrastructure Development

  • China is rapidly expanding its clean energy resources which support data center operations crucial for powering advanced computing needs related to AI development.

Strategic Location Choices

  • Data centers are being built in western provinces where renewable energy sources are abundant despite potential latency issues due to distance from urban centers.

Competitive Edge: Speed vs. Quality

Benchmarking Performance

  • Current estimates suggest that Chinese models lag behind American counterparts by 3–9 months depending on specific benchmarks and evaluations.

Focus Beyond Cutting-edge Models

  • There’s an emphasis on integrating strong models into practical applications rather than solely achieving top-tier performance metrics.

Public Sentiment Towards AI

Anxiety Around Technological Advancement

  • In contrast to American fears about job displacement due to automation, public sentiment in China revolves around concerns about falling behind technologically.

Integration Over Resistance

  • Individuals express a desire for greater integration of technology into their lives rather than resisting it; this reflects broader national anxieties regarding competitiveness.

Labor Market Dynamics

Youth Employment Concerns

  • With high youth unemployment rates at approximately 17%, there’s pressure among new graduates who fear they may not secure desirable jobs without technological proficiency.

Shifting Perspectives on Welfare

  • Discussions around welfare state adaptations due to potential job displacement from AI have begun emerging among policymakers amid changing labor market dynamics.

Social Implications of Technology

Governmental Concerns About Social Effects

  • The Chinese government has expressed worries about how technology impacts social life; previous crackdowns indicate a proactive approach towards perceived nonproductive activities such as gaming or excessive use of digital companions.

Understanding China's Perspective on AI and AGI Development

China's Attitude Towards Technology and Super Intelligence

  • The Chinese may perceive U.S. tech companies as overly confident, believing that a return to super intelligence is unlikely, allowing them to remain 3-6 months behind while still catching up.
  • Alternatively, China might be concerned about super intelligence and is strategizing contingency plans if the U.S. advances significantly ahead in AI technology.

Speculations on Secret AGI Projects

  • There are questions about whether China could be conducting a covert AGI project without U.S. awareness; however, it is believed that such large-scale operations would likely be detected.
  • Current monitoring of China's nuclear build-out suggests that any significant secret projects related to AGI would not go unnoticed due to the required scale of resources.

Divergent Views on Technology Integration

  • China seems to accept its current position in the AI race rather than attempting to secretly outpace the U.S., viewing technology through a different lens focused on practical applications like education and healthcare.
  • A thought experiment considers what signs would indicate a secret AGI program in China, particularly regarding their interest in advanced Nvidia chips.

Semiconductor Strategy and Long-Term Planning

  • Despite recent relaxations in export controls for Nvidia chips, China has opted not to acquire these chips quickly, indicating a long-term strategy focused on developing its semiconductor capabilities instead of immediate gains.

Espionage and AI Model Distillation Practices

  • Some Chinese labs appear to engage in distillation practices by training weaker models using outputs from stronger American models without authorization, raising concerns about intellectual property rights.
  • This method does not equate to outright IP theft but involves unauthorized use of existing models for improvement.

Risks Associated with AI Development

  • The potential for rapid advancements in AI could lead to scenarios where being slightly behind becomes detrimental; this raises questions about geopolitical responses from China if they feel threatened by U.S. advancements.

Geopolitical Implications of an Arms Race

  • If the U.S. were significantly ahead in AI development, there are speculations about whether China might take aggressive actions like invading Taiwan as a desperate measure.

Policy Recommendations for the United States

  • The discussion emphasizes stepping back from an all-out race mentality towards more measured policies regarding competition with China over AI development.

Balancing Safety and Progress

  • Concerns arise that focusing solely on competing with China's perceived threat may lead to reckless policies that neglect necessary safety regulations around AI technologies.

Open Source vs Commercial Models

  • There's an argument for adopting strategies similar to China's focus on integration and deployment rather than just racing towards cutting-edge developments; open-source models could enhance global adoption.

Export Controls Debate

  • The complexities surrounding chip exports highlight both short-term benefits of slowing down Chinese advancements while potentially accelerating their self-reliance in semiconductor production over time.

Cybersecurity Risks vs AGI Threat Perception

  • While concerns exist regarding catastrophic risks associated with AGI, there’s recognition that cyber threats pose more immediate dangers which need addressing alongside regulatory considerations.

Future Engagement Strategies

  • Engaging with China regarding mutual interests in AI safety is suggested as essential despite low expectations for successful negotiations or binding agreements at this stage due to trust issues between nations.
Video description

The United States and China are really the only two countries that matter right now in shaping the A.I. future. As President Trump and President Xi Jinping meet in Beijing, there’s a kind of Cold War atmosphere, with people talking about an A.I. arms race. But who is winning? Are we even in a race at all? Kyle Chan, a foreign policy fellow at the Brookings Institution, says it’s hard to call it a race because the U.S. and China have very different A.I. goals. 00:00:13 U.S. vs. China in A.I. 00:03:30 Everyday A.I. in China 00:07:30 The trillion dollar chip question 00:12:10 China's A.I. advantage 00:15:30 China's public mood on A.I. 00:19:10 A.I., job displacement and social concerns 00:22:10 Robots for China's labor shortage 00:26:00] China's view on America's AGI fixation 00:31:00 Distilling A.I. models 00:38:50 U.S. needs more A.I. deployment 00:42:20 U.S. chip policy and the hawk's argument Thoughts? Email us at interestingtimes@nytimes.com. Read the full transcript here: https://www.nytimes.com/2026/05/14/opinion/china-trump-ai-xi.html Unlock full access to New York Times podcasts and explore everything from politics to pop culture. Subscribe today at nytimes.com/podcasts or on Apple Podcasts and Spotify.