Big Technology Podcast: Why China's AI Strategy Just Might Work — With Grace Shao

Host: Alex Kantrowitz
Guest: Grace Shao, Hong Kong-based Writer and Analyst
Release Date: April 16, 2025

1. Introduction and Setting the Stage

In this episode of the Big Technology Podcast, host Alex Kantrowitz delves into the rising prominence of China's artificial intelligence (AI) strategy. With China's Deep Seek breakthrough, questions arise: Is Deep Seek an isolated success, or does it signify a broader shift positioning China as a formidable global leader in AI? To explore these questions, Alex welcomes Grace Shao, a renowned writer and analyst specializing in China's tech landscape.

2. Deep Seek: Outlier or Indicator of Broader Strength? (01:29 - 04:22)

Alex Kantrowitz opens the discussion by asking whether Deep Seek is merely an exceptional case or a harbinger of China's growing competitiveness in AI.

Grace Shao responds firmly, stating, "I think definitely we're seeing that it's not an outlier" (01:55). She elaborates on how Deep Seek has revolutionized perceptions of scaling laws in AI, bolstering confidence within China's AI entrepreneurial sphere. Shao highlights the emergence of a new generation of Chinese AI entrepreneurs, predominantly born in the '80s and '90s, who are mission-driven rather than solely profit-oriented. This shift contrasts sharply with the earlier, more financially motivated entrepreneurs like Jack Ma of Alibaba.

Additionally, Grace points out the influence of geopolitical tensions over the past few years, fostering a nationalistic drive to lead in global innovation. She concludes, "we can continue to watch. And I expect that we'll see more and more innovation in AI space coming out of China" (04:22).

3. China's Open Source AI Strategy: A Strategic Move (05:50 - 10:17)

Alex probes into China's strategy of open-sourcing AI models, questioning whether it's a tactic to undermine U.S. AI research labs by eliminating the perceived moat of proprietary AI.

Grace affirms this by explaining China's pragmatic approach. She notes, "China actually... went with a AI monetization strategy that's really heavily leaning toward consumer versus enterprise" (06:22). Grace attributes this to the low adoption of Software as a Service (SaaS) in China, driven by an economy still heavily reliant on manufacturing rather than knowledge-based industries. This environment discourages enterprise software proliferation and instead promotes consumer-facing applications.

She further explains that by open-sourcing models like Deep Seek, Chinese companies encourage developers to build upon these foundations, fostering a robust ecosystem of AI-powered consumer products. Grace summarizes, "100%. I think you put it in a very succinct way for me" (10:17), agreeing with Alex's assessment of China's strategic focus on consumer applications leveraging open-source AI.

4. Productization of AI: Consumer vs. Enterprise Focus (11:15 - 17:57)

Alex draws parallels between Chinese and American tech strategies, citing Mark Zuckerberg's approach with Meta as mirroring China's open-source AI model to enhance consumer-facing products like Messenger and Instagram.

Grace concurs, emphasizing that while companies like Tencent have integrated Deep Seek into existing platforms like WeChat to leverage their vast user base, others like ByteDance struggled with standalone AI applications due to average performance. She states, "Once Tencent integrated Deep Seek into their WeChat product, it was like game over for them" (13:38), highlighting the competitive edge gained through integration rather than independent app development.

Grace further explains that Alibaba and other Chinese tech giants are focusing on embedding AI into their infrastructure layers, intending to monetize through applications and user engagement rather than direct sales of AI models. This strategy contrasts with the U.S., where AI products are often geared towards productivity and efficiency for knowledge workers.

5. Broader AI Innovations: Beyond Language Models (17:57 - 25:18)

Alex inquires about other areas where China is making strides in AI beyond language models like Deep Seek and Alibaba's Quen.

Grace broadens the discussion, highlighting China's advancements in physical AI, particularly robotics. She notes, "China is definitely in this" (19:54), referring to the integration of AI into manufacturing and robotics sectors. Grace credits China's robust manufacturing infrastructure, affordable labor, and comprehensive supply chains for enabling rapid development and deployment of AI-powered devices.

She also touches upon the emergence of AI in traditional sectors, such as Traditional Chinese Medicine (TCM), where AI is being used to enhance diagnostic processes. Grace remarks, "they almost use it as like a fun entertainment tool" (19:53), illustrating the diverse applications of AI across various industries in China.

6. Robotics in China: Leading the Charge (25:24 - 48:26)

Alex shifts focus to robotics, praising the sophistication and flexibility of Chinese robots showcased on platforms like Twitter, which surpass their Western counterparts in functionality.

Grace elaborates on China's dominance in the robotics market, particularly in humanoid and industrial robots. She cites Unitree Robotics, which holds "about 70% of global sales of humanoid robots" (48:26), and highlights the affordability of these robots compared to Western models like Boston Dynamics. Grace attributes this success to China's deep expertise in mechatronics—a synergy of mechanical engineering and software—that stems from decades of manufacturing prowess.

Moreover, she discusses the challenges of regulation and safety standards that currently hinder the mass adoption of robotics in consumer households, emphasizing the need for robust safety protocols as these technologies evolve.

7. The Six Dragons of Hangzhou: An Innovation Hotspot (53:04 - 57:54)

Alex asks about the "Six Dragons" emerging from Hangzhou, a city instrumental in China's AI and tech advancements.

Grace describes Hangzhou as a vibrant second-tier city, home to major conglomerates like Alibaba, NetEase, and Geely. She explains that Hangzhou's strategic location—close to Shanghai yet distinct from Beijing—fosters an environment conducive to innovation and entrepreneurship. The city's rich ecosystem, bolstered by strong STEM education from institutions like Jiujiang University, has birthed numerous AI-focused companies, including Game Science Mani Many Core, BrainCo, and Deep Robotics.

Grace emphasizes that Hangzhou's supportive local government policies and its blend of traditional manufacturing with cutting-edge tech make it a fertile ground for AI startups and innovations, further solidifying China's position in the global AI landscape.

8. AI Agents and Manus: Evaluating Emerging Technologies (58:43 - 60:52)

Alex brings up the emergence of AI agents like Manus, questioning whether such developments signify a new phase akin to Deep Seek's impact.

Grace provides a measured perspective, acknowledging that while AI agents like Manus are intriguing, they do not yet represent a transformative breakthrough comparable to Deep Seek. She notes, "It's not a deep seek moment" (59:03), suggesting that these agents are still in their nascent stages and have not fundamentally altered the AI development landscape. Grace also touches on the challenges of accessibility and exclusivity surrounding products like Manus, indicating that widespread adoption and impact remain to be seen.

9. Overcoming Challenges: Export Controls and Chip Shortages (61:14 - 62:45)

Alex addresses the significant hurdles Chinese AI companies face, particularly export controls and GPU shortages, which could impede their global competitiveness.

Grace acknowledges these challenges, highlighting that despite China's extensive data resources and innovative capabilities, the scarcity of high-quality chips remains a critical bottleneck. She mentions that "Deep Seek trained on the H8 hundreds" to circumvent chip bans (61:14), but emphasizes that without breakthroughs in domestic chip production from companies like Huawei or Baidu, Chinese AI advancements could be stymied. Grace suggests that overcoming these hardware limitations is essential for China to sustain and enhance its AI leadership on the global stage.

10. Government Support and Future Outlook (36:59 - 43:49)

Alex references recent governmental support for AI during China's National People's Congress, noting a shift towards proactive tech policies.

Grace reinforces the significance of this support, stating, "the government is very supportive of AI entrepreneurship" (40:48). She explains that municipal and provincial policies, especially in tech hubs like Zhejiang and Guangdong, provide critical incentives for AI development. Grace also touches upon the resilient entrepreneurial spirit in China, which has adapted post-regulatory clampdowns and geopolitical tensions, positioning AI innovation as a cornerstone for economic revival and growth.

Conclusion

In this comprehensive discussion, Grace Shao elucidates why China's AI strategy is gaining traction and why it might just work. From leveraging open-source models and integrating AI into consumer products to leading in robotics and overcoming infrastructural challenges, China's multifaceted approach positions it as a formidable force in the global AI arena. However, persistent challenges like chip shortages and regulatory hurdles remain critical factors that will influence the trajectory of China's AI ambitions.

Notable Quotes:

• "I think definitely we're seeing that it's not an outlier." — Grace Shao (01:55)
• "100%. I think you put it in a very succinct way for me." — Grace Shao (10:17)
• "Once Tencent integrated Deep Seek into their WeChat product, it was like game over for them." — Grace Shao (13:38)
• "China is definitely in this." — Grace Shao (19:54)
• "I think it's not easy." — Grace Shao (61:14)

For more insights and in-depth analysis, subscribe to Grace Shao's newsletter AI Proem.